CA2515470C - Methods and apparatus for managing computing deployment in presence of variable workload - Google Patents
Methods and apparatus for managing computing deployment in presence of variable workload Download PDFInfo
- Publication number
- CA2515470C CA2515470C CA2515470A CA2515470A CA2515470C CA 2515470 C CA2515470 C CA 2515470C CA 2515470 A CA2515470 A CA 2515470A CA 2515470 A CA2515470 A CA 2515470A CA 2515470 C CA2515470 C CA 2515470C
- Authority
- CA
- Canada
- Prior art keywords
- servers
- workload level
- future workload
- causing
- insufficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000009471 action Effects 0.000 claims abstract description 68
- 230000004044 response Effects 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 12
- 238000012544 monitoring process Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 230000002567 autonomic effect Effects 0.000 abstract description 7
- 238000007726 management method Methods 0.000 description 19
- 230000003466 anti-cipated effect Effects 0.000 description 9
- 238000013459 approach Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000013439 planning Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000013068 supply chain management Methods 0.000 description 2
- 230000003442 weekly effect Effects 0.000 description 2
- 210000003403 autonomic nervous system Anatomy 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 238000004883 computer application Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004043 responsiveness Effects 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 238000004513 sizing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/008—Reliability or availability analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
Abstract
Automated or autonomic techniques for managing deployment of one or more resources in a computing environment based on varying workload levels. The automated techniques may comprise predicting a future workload level based on data associated with the computing environment. Then, an estimation is performed to determine whether a current resource deployment is insufficient, sufficient, or overly sufficient to satisfy the future workload level. Then, one or more actions are caused to be taken when the current resource deployment is estimated to be insufficient or overly sufficient to satisfy the future workload level. Actions may comprise resource provisioning, resource tuning and/or admission control.
Description
METHODS AND APPARATUS FOR MANAGING COMPUTING
DEPLOYMENT IN PRESENCE OF VARIABLE WORKLOAD
Cross Reference to Related Application The present application is related to U.S. Patent 7,039,559 entitled "Methods and Apparatus for Performing Adaptive and Robust Prediction," filed concurrently herewith.
Field of the Invention The present invention relates generally to management of computing systems or networks and, more particularly, to techniques for managing computing deployment associated with such a system or network in the presence of variable workload.
Background of the Invention An important challenge in managing deployments of computing resources in a computing system or network is dealing with variable traffic. For instance, in a computing system or network associated with the World Wide Web or Internet, it is important to have sufficient computing resources (e.g., web servers, application servers, transaction/database servers) supporting a web site to ensure that the end-user experience is not compromised (e.g., by slow response time), even when the web site is under heavy load with respect to the utilization of one or more applications executed in association with the web site. As is known, an application generally refers to a one or more computer programs designed to perform one or more specific functions, e.g., supply chain management.
One approach to sizing a deployment supporting a particular application is to estimate the anticipated workload traffic pattern, and use enough resources to accommodate the peak anticipated load, using capacity planning approaches.
This static arrangement can result in significant resource under-utilization since most workload traffic is quite variable, e.g., with marked diurnal, weekly, etc., patterns.
A refinement on the above approach is to do scheduled or planned source reallocation based on a long-term (e.g., one to several days) forecast of anticipated traffic.
This approach is also often inadequate as it relies on the accuracy of a long-term forecast (which may, e.g., underestimate the success of a sales promotion) and is also exposed to unanticipated events (e.g., a traffic surge at news web sites such as experienced at CNN's web site on 9/11/01).
Another key disadvantage of existing computing deployment approaches is that they generally require some form of manual intervention, e.g., via expert operators, to adjust for resource imbalance.
Accordingly, it would be desirable to have automated or autonomic techniques for managing a computing deployment, associated with a computing system or network, which handle variable workload more efficiently and effectively than existing approaches.
Summary of the Invention The present invention provides automated or autonomic techniques for managing a computing deployment, associated with a computing system or network, which handle variable workload more efficiently and effectively than existing approaches.
In one aspect of the invention, techniques are provided for managing deployment of one or more resources in a computing environment based on varying workload levels.
The techniques may comprise predicting a future workload level based on data associated with the computing environment. Then, an estimation is performed to determine whether a current resource deployment is insufficient, sufficient, or overly sufficient to satisfy the future workload level. Then, one or more actions are caused to be taken when the current resource deployment is estimated to be insufficient or overly sufficient to satisfy the future workload level. Actions may comprise resource provisioning, resource tuning and/or admission control Advantageously, the present invention may provide for proactively maintaining a service level objective, such as response time, for a computing deployment in the face of variable workload. In particular, by making changes to a computing deployment in an automated or autonomic fashion, the techniques employed by the invention are effective at accommodating unanticipated workload.
The present invention also advantageously provides a methodology for an application owner to attempt to ensure satisfaction of one or more service objectives associated with the execution of an application that is hosted by a service provider. This may be accomplished by the application owner contracting with the service provider to host the application and to implement a computing deployment management system as provided herein.
Accordingly, in one aspect there is provided an automated method of managing deployment of a plurality of servers in a computing environment based on varying workload levels, the method comprising the steps of:
predicting a future workload level based on data associated with the computing environment;
wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level, by a processor in response to instructions stored on a non-transitory computer readable medium, as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to another aspect there is provided an apparatus for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the apparatus comprising:
a memory; and at least one processor coupled to the memory and operative to:
(i) predict a future workload level based on data associated with the computing environment using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
(iii) cause one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and (iv) selectively adapt the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing operation further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing operation further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to yet another aspect there is provided an article of manufacture for managing deployment of a plurality of servers in a computing environment based on varying workload levels, comprising a computer readable storage medium containing one or more programs which when executed implement the steps of predicting a future workload level based on data associated with the computing environment, wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
3a estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to still yet another aspect there is provided an automated system for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the system comprising:
a solution manager module comprising memory and at least one processor coupled thereto and operative to:
(i) predict a future workload level based on data associated with the computing environment, wherein the prediction operation further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level; and (iii) selectively adapt the forecast horizon as a function of a time needed to effectuate at least one of one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and a deployment manager coupled to the solution manager module, comprising a memory and at least one processor coupled thereto and operative to:
(i) provide the data associated with the computing environment to the solution manager module; and 3b (ii) effect the one or more actions to be taken, in response to the solution manager module, when the current deployment of servers is estimated by the solution manager module to be one of insufficient and overly sufficient to satisfy the future workload level, wherein the one or more actions to be taken comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient and wherein the one or more actions to be taken comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient;
wherein the deployment manager further comprises:
(i) a monitoring module for providing access to workload data;
(ii) a provisioning module for performing resource provisioning;
(iii) a tuning interface module for changing one or more configuration parameters associated with one or more of the servers; and (iv) a throttling interface module for causing a manipulation of one or more admission queues on one or more of the servers.
These and other objects, features and advantages of the present invention will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in connection with the accompanying drawings.
Brief Description of the Drawings FIG. 1 is a block diagram illustrating a computing deployment management system according to an embodiment of the present invention and an overall environment in which such system may operate;
FIG. 2 is a flow diagram illustrating a computing deployment management methodology according to an embodiment of the present invention;
FIG. 3 is a graphical representation illustrating performance of a computing system or network in accordance with principles of the present invention; and FIG. 4 is a block diagram illustrating a generalized hardware architecture of a computer system suitable for implementing a computing deployment management system according to the present invention.
3c Detailed Description of Preferred Embodiments The present invention will be explained below in the context of an illustrative web-based computing network environment. That is, the computing resources being managed (e.g., application servers, database connections, input/output paths, etc.) are associated with one or more web sites. However, it is to be understood that the present invention is not limited to such a particular environment. Rather, the invention is more generally applicable to any computing environment in which it is desirable to automatically or autonomically manage and compute resource deployment in the face of variable workload.
As is known, "autonomic" computing generally refers to a comprehensive and holistic approach to self-managed computing systems with a minimum of human interference, see, e.g., P. Horn, "Autonomic Computing IBM's Perspective on the State of Information Technology," IBM Research, October 2001. The teiin derives from the body's autonomic nervous system, which controls key functions without conscious awareness or involvement.
More specifically, one of the goals of autonomic computing is to automate some or all of the tasks an expert operator or administrator would typically carry out. Thus, as will be appreciated from the inventive principles presented herein, the computing deployment techniques of the invention are able to operate automatically or autonomically.
Referring initially to FIG. 1, a block diagram illustrates a computing deployment management system according to an embodiment of the present invention and an overall environment in which such system may operate. As shown, the environment 100 comprises a computing deployment management system 100. The computing deployment management system 100, itself, comprises a solution manager 120 and a deployment manager 130. The solution manager 120, itself, comprises a control logic engine 122, a forecaster module 124, a performance estimator module 126, an admission control module 128, and a tuning module 129. The deployment manager 130, itself comprises a monitoring module 132, a provisioning module 134, a tuning interface module 136, and a throttling interface module 138.
Further, as shown, the environment 100 comprises an application level 140. The application level, itself, comprises resource pool 142 (resource pool A
comprising, for example, application servers), resource pool 144 (resource pool B comprising, for example, database connections), and resource pool 146 (resource pool C
comprising, for example, input/output paths).
Accordingly, the architecture shown in FIG. 1 is organized into three levels:
(a) the application level (denoted as 140) and associated resources on which the application can be deployed; (b) a deployment management level (denoted as 130) which provides connection and control of resources; and (c) a solution management level (denoted as 120) which performs the real-time analysis and planning required to initiate actions that are required to maintain a service level objective. These three levels are discussed in further detail below.
Application deployment typically requires a mix of resources of various types, such as, for example, an HTTP (hypertext transport protocol) server, an application server, a database server, storage, connections, I/O paths, etc. In a typical computing deployment (e.g., a data center), these resources could be available from a managed pool.
FIG. 1 illustrates three such managed resource pools 142, 144 and 146. It is understood that, depending on the application, a predetermined number of each of the resources, sufficient to satisfy anticipated workloads, is available for use in the managed resource pools. It is to be further understood that while FIG. 1, and the above description, mention certain resources, the invention is not limited to any particular resources. Rather, the invention may manage any and all types of resources including, but not limited to, hardware components, software components, and combinations thereof It is to be understood that a resource may also be an application, itself, or some portion thereof.
The deployment manager 130 interfaces with relevant resources of the application level 140 to monitor measurement/configuration data (e.g., through resource-dependent sensors such as, for example, response time probes, vmstat data from the operating system such as Unix, snapshot data from a database such as IBM Corporation's DB2, or through custom interfaces or standard interfaces implemented using the Common Information Model) and to control the resources (e.g., through resource-dependent effectuators such as, for example, the node agent on an application server such as IBM
Corporation's WebSphere Application Server, the application programming interface for changing configuration parameters in a database such as IBM Corporation's DB2). Hence, the deployment manager is able to perform resource provisioning (via provisioning module 134) which, by way of example, for a piece of hardware, can range from: (i) deploying an operating system on a computer without an installed operating system, e.g., an x86 system on which Windows or Linux can be installed, or replacing an existing operating system on a computer with a new operating system; (ii) deploying appropriate middleware on top of the operating system; (iii) deploying an application with associated data; and (iv) performing relevant cluster management/federation to enable an added resource to support the application. Advantageous features of this provisioning capability include not only rapidly and automatically adding resources when needed, for example, in response to an unexpected workload surge, but also removing resources when no longer needed, hence minimizing the greater cost of additional resources.
In addition, the deployment manager 130 (via tuning interface module 136) resets resource configuration parameters (e.g., memory pool sizes such as buffer pools for a database, ORB (object request broker) thread pool size in an application server) which is important for resource tuning. Resource tuning generally refers to the technique of changing one or more configuration parameters associated with a resource in a manner which helps achieve a goal such as minimizing response time or maximizing throughput.
The deployment manager 130 (via throttling interface module 138) also manipulates admission queues on the resources (e.g., for admission control/request throttling).
Throttling generally refers to rejecting incoming requests based on some policies that identify service classes such as, for example, type of request (buy versus browse at an e-commerce site which can be distinguished by a uniform resource locator), origin (preferred customers), etc. By rejecting such requests, the incoming load to a computing deployment may be reduced to a manageable level. The deployment manager (via monitoring module 132) also provides access to the workload data (e.g., throughput, response time, etc.).
The solution manager 120 is responsible for maintaining the service objective for the particular application deployment. "Service objective" may refer to requirements and/or preferences specified in accordance with a service level agreement (SLA). That is, by way of example, such service objectives may deal with how service applications are hosted at a third party infrastructure, while ensuring a certain level of end-client satisfaction. As is known, businesses increasingly run their applications using infrastructure (e.g., server, network connectivity) provided by a third party, generally referred to as the "service provider." Many companies, such as IBM Global Services, host web sites and/or provide other computer hosting services. An SLA provides a means by which the expectations of the service provider can be negotiated with the customer. An SLA between an application owner and the service provider defines terms and conditions for this hosting service. The SLA may, for example, include expected response time, bandwidth throughput at the network and/or servers, disk space utilization, availability, i.e., up-time of network and server resources, as well recovery time upon failure, and pricing for various levels of service. However, it is to be appreciated that a service objective does not have to come from an SLA, which typically has legal consequences. A
service level objective can often be negotiated within an enterprise, e.g., between the information technology (IT) department and the purchasing department for whom they may be deploying an online purchase order system. Also an e-commerce site or even a , , place like Google may want to maintain a good service level, with regard to something like response time, so that the user experience is good.
Accordingly, so as to sufficiently maintain the particular service objectives, the solution manager 120, in accordance with the control logic engine 122, decides: (i) when action needs to be taken; and (ii) what action to take. The control logic engine 122 accomplishes these tasks, as will be explained below, in accordance with forecaster module 124, performance estimator module 126, admission control module 128, and tuning module 129. That is, it is to be understood that the control logic engine (CLE) 122 serves as a controller for the functions provided by the other modules in the solution manager 120. It is to be understood, however, that the functional arrangement shown in block 120 is illustrative in nature and, thus, other arrangements for controlling the functionality provided by the solution manager may be employed within the scope of the principles of the present invention.
One advantageous feature of the solution manager is the ability to proactively manage the service objective, hence, minimizing/avoiding a possible violation.
This is accomplished by a combination of forecasting (via forecaster module 124) on the workload (obtained from the monitoring component in the deployment manager) in combination with performance estimation (via performance estimator 126) to check if a service level violation is anticipated. In a preferred embodiment, the control logic component 122 is essentially a rule based component that, based on the forecasting and performance estimation results, decides which action or actions to take.
Examples of actions that can be taken include provisioning (adding/removing resources), admission control, and resource tuning.
Resource tuning may be effective when the workload changes. For instance, in a e-commerce site, the relative buy to browse mix can change, and since these transactions can draw on somewhat different resources in a database, e.g., such as use different buffer pools, or require more sorting, the overall system responsiveness may be improved by changing database configuration parameters, such as relative buffer pool sizes, or sort heaps. This is determined and accomplished by the tuning module 129, via the tuning interface 136.
Admission control actually rejects some of the incoming requests based on some policy, e.g., admit only preferred customers, or only buy transactions, such that the system is not overloaded. This kind of action may be taken if others actions such as provisioning or resource tuning cannot achieve the desired results.
In any case, the magnitude of the action (e.g., how many servers to add, how much to change a configuration parameter) is determined using the performance estimation capability. These actions could be taken separately or in combination.
The following is an illustrative description of a functional operation of a computing deployment management system according to the present invention. In this example, one type of control action is considered, namely, application provisioning, which is accomplished by server addition to/removal from the active cluster.
Referring now to FIG. 2, a flow diagram illustrates a computing deployment management methodology 200 according to an embodiment of the present invention.
The solution manager 120 keeps track of the workload (transaction rate), service objective (response time), and configuration (resources which are active, idle, and in transition states) that the control logic engine (CLE) 122 obtains from the monitoring module 132 in the deployment manager 130 (step 202).
Based on monitored data such as the transaction rate history, the solution manager 120 uses the forecaster module 124 to project into the future the expected workload (step 204).
The forecast can be long-term (e.g., hours/days) or short-term (e.g., seconds/minutes). To deal with unanticipated changes in the workload (e.g., outside of normal daily/weekly, etc.) variations, the forecaster is preferably adaptive and rapidly learns from recent history. The forecast horizon does not need to be long, e.g., the horizon may be approximately the amount of time needed for the resource action (application provisioning in this situation) to take effect.
In a preferred embodiment, the forecaster module 124 may be implemented using the techniques described in U.S. Patent 7,039,559 entitled "Methods and Apparatus for Performing Adaptive and Robust Prediction," filed concurrently herewith.
However, it is to be appreciated that the forecaster module 124 may employ other forecasting techniques such as, for example, those described in "Time Series Analysis: Forecasting and Control,"
by G.E.P. Box et al. (rev. ed.), or those employed in the Hyperion Forecasting Suite available from Hyperion.
The solution manager 120 then uses the performance estimator 126 to check if the current resource deployment is insufficient, sufficient (i.e., adequate) or overly sufficient (i.e., excessive) for maintaining the service objective threshold (e.g., response time) based on the recent and forecasted workload traffic (step 206). While the invention is not limited to any particular performance estimation techniques, examples of such techniques that may be employed include those described in "Configuration and Capacity Planning for Solaris Servers," by Brian L Wong, and "Capacity Planning for Web Services:
Metrics, Models, and Methods" by Daniel A. Menasce et al., and/or those referred to as PATROL Perfom and PATROL Predict from BMC Software, and High Volume Web Site Performance Simulator for Websphere from IBM Corporation. Based on recommendations of the performance estimator, and knowledge of state and number servers in the relevant resource pool, the solution manager 120 sends a request to the application provisioner 134 in the deployment manager 130 to add/remove the appropriate number of servers, if necessary (step 208). That is, if it is determined by the CLE 122 that the current resources are sufficient, then no provisioning may be necessary at this time.
, If provisioning is needed (step 210), the provisioner 134 acts rapidly to add a server from the idle state to the active cluster running the application.
Rapid addition (e.g., from about several minutes to less than about one minute) may be achieved by a combination of loading saved application images on the server instance being added, starting the server, and allowing the cluster manager to send requests to it.
Note that speed of resource addition may be increased if the servers are preloaded with the application image. Hence, the preloaded servers need only to be started and activated within the cluster. To remove the server from the active cluster, the provisioner 134 disables new incoming requests to the server from the cluster manager, while allowing existing work to continue. This allows a natural quiescing of the work on that server, which is then stopped.
It is to be appreciated that by use herein of the phrases "insufficient" and "overly sufficient" (and, similarly, "inadequate" and "excessive") with respect to computing resources, it is generally not only meant that there are not enough (insufficient) resources deployed or too many (overly sufficient) resources deployed, but also that one or more resources are improperly configured to meet the anticipated workload (e.g., due to a workload mix change, etc.) or that certain admission requests that should not be admitted are being admitted Thus, in accordance with the invention, the resource deployment may be made "sufficient" (or, similarly, "adequate") not only by adding or removing resources, but also by tuning one or more resources and/or throttling certain admission requests.
The following description provides some results from an exemplary implementation of the management system described herein. In the exemplary deployment considered, the application is a supply chain management. The application is run on a cluster of web application servers with a database server in the back-end. In this example, the system is faced with a large unanticipated surge of incoming transactions requests (doubling every minute to about 20 times the normal load) while trying to stay below a response time service objective of two seconds. The prediction horizon of the short term forecaster is one minute, since the rapid application deployment capability of the provisioner is about 30 to 40 seconds.
The performance of the system is shown in FIG. 3. The prediction of the transaction rate (top panel of FIG. 3, line 302) initially lags the surge in the actual transaction rate (top panel of FIG. 3, line 304), but very quickly catches up and is able to provide useful guidance on the anticipated transaction rate. This is in turn used to calculate the requisite number of application servers using the performance estimator (126 of FIG. 1), and the servers are quickly brought into active (middle panel of FIG. 3, line 306) duty with the rapid provisioning capability of the application provisioner (134 of FIG. 1). The amount of time a server is in the transitional starting state (middle panel of FIG. 3, line 308) while going from idle to active is only about 30 seconds.
The combination of the forecasting and performance estimation in the solution manager together with the rapid application provisioning capability of the deployment manager allows the system to stay below the two second response time target (bottom panel of FIG. 3, line 310) in the face of an extremely aggressive surge. After the surge is over, the servers are returned to the pool.
Referring now to FIG. 4, a block diagram illustrates a generalized hardware architecture of a computer system suitable for implementing a computing deployment management system according to the present invention. For instance, the functional components shown in FIG. 1 with respect to the solution manager 120 and the deployment manager 130 may be implemented on one or more computer systems of the type shown in FIG. 4. Of course, separate functional components may be implemented on their own dedicated computer system. However, it is to be appreciated that the computing deployment management system of the invention is not intended to be limited to any particular computer platform, arrangement or implementation.
In this illustrative implementation 400, a processor 402 for implementing management methodologies and functionalities of the invention as described herein is operatively coupled to a memory 404 and input/output (I/0) devices 406, via a bus 408 or an alternative connection arrangement. It is to be appreciated that the term "processor" as used herein is intended to include any processing device, such as, for example, one that includes a central processing unit (CPU) and/or other processing circuitry (e.g., digital signal processor (DSP), microprocessor, etc.). Additionally, it is to be understood that the term "processor" may refer to more than one processing device, and that various elements associated with a processing device may be shared by other processing devices.
The term "memory" as used herein is intended to include memory and other computer-readable media associated with a processor or CPU, such as, for example, random access memory (RAM), read only memory (ROM), fixed storage media (e.g., hard drive), removable storage media (e.g., diskette), flash memory, etc. The memory may preferably be used to store data and computer programs associated with the invention.
In addition, the term "I/O devices" as used herein is intended to include one or more input devices (e.g., keyboard, mouse) for inputting data to the processing unit, as well as one or more output devices (e.g., CRT display) for providing results associated with the processing unit.
It is to be appreciated that the methodologies of the present invention are capable of being implemented in the form of computer readable media. The term "computer readable media" as used herein is intended to include recordable-type media, such as, for example, a floppy disk, a hard disk drive, RAM, compact disk (CD) ROM, etc., as well as transmission-type media.
Accordingly, one or more computer programs, or software components thereof, including instructions or code for performing the methodologies of the invention, as described herein, may be stored in one or more of the associated storage media (e.g., ROM, fixed or removable storage) and, when ready to be utilized, loaded in whole or in part (e.g., into RAM) and executed by the processor 402.
In any case, it is to be appreciated that the techniques of the invention, described herein and shown in the appended figures, may be implemented in various forms of hardware, software, or combinations thereof, e.g., one or more operatively programmed general purpose digital computers with associated memory, implementation-specific integrated circuit(s), functional circuitry, etc. Given the techniques of the invention provided herein, one of ordinary skill in the art will be able to contemplate other implementations of the techniques of the invention.
Accordingly, as explained herein in detail, the present invention provides an architecture which enables automated proactive management of a system in the face of variable workload or unexpected workload variability. That is, a system for proactive management of computer applications that experience unexpected variability in demand is provided which may comprise components for forecasting, performance modeling, control, and reconfiguration. The forecasting is used to anticipate future workload and actions are taken sufficiently rapidly to accommodate the anticipated workload, while reducing exposure to potential violations of a service level objective.
Actions are taken in a cost-effective manner that minimizes resource under-utilization while avoiding excessive repeated resource configuration changes, i.e., oscillations. Actions taken are based on models/estimations of performance based on available resources together with anticipated workload and consideration of the service objective. Actions may include resource provisioning and/or resource tuning and/or admission control.
An architecture of the invention may be structured into: (a) an application to be managed; (b) a deployment manager that provides a generic interface to that system for sensing and effecting control; and (c) a solution manager that determines when actions are to be taken based on this generic interface to monitoring, and requests the necessary actions through the generic control interface. The deployment manager can manage multiple configurations, allowing for rapid deployment or redeployment of resources as a result of being able to: (a) take actions specific to a particular resource node, a particular configuration, or particular function; and (b) automatically set application appropriate database paths and parameters.
As is further evident, the invention further provides techniques operating an adaptive system in which: (a) workload and service level metrics are collected; (b) future values of workload measurements are forecast; (c) these forecasted values and the desired service level objective are used to determine the actions required to ensure adequate service levels; and (d) the actions themselves are initiated in a way that minimizes oscillations and yet responds promptly.
Although illustrative embodiments of the present invention have been described herein with reference to the accompanying drawings, it is to be understood that the invention is not limited to those precise embodiments, and that various other changes and modifications may be made therein by one skilled in the art without departing from the scope of the invention.
DEPLOYMENT IN PRESENCE OF VARIABLE WORKLOAD
Cross Reference to Related Application The present application is related to U.S. Patent 7,039,559 entitled "Methods and Apparatus for Performing Adaptive and Robust Prediction," filed concurrently herewith.
Field of the Invention The present invention relates generally to management of computing systems or networks and, more particularly, to techniques for managing computing deployment associated with such a system or network in the presence of variable workload.
Background of the Invention An important challenge in managing deployments of computing resources in a computing system or network is dealing with variable traffic. For instance, in a computing system or network associated with the World Wide Web or Internet, it is important to have sufficient computing resources (e.g., web servers, application servers, transaction/database servers) supporting a web site to ensure that the end-user experience is not compromised (e.g., by slow response time), even when the web site is under heavy load with respect to the utilization of one or more applications executed in association with the web site. As is known, an application generally refers to a one or more computer programs designed to perform one or more specific functions, e.g., supply chain management.
One approach to sizing a deployment supporting a particular application is to estimate the anticipated workload traffic pattern, and use enough resources to accommodate the peak anticipated load, using capacity planning approaches.
This static arrangement can result in significant resource under-utilization since most workload traffic is quite variable, e.g., with marked diurnal, weekly, etc., patterns.
A refinement on the above approach is to do scheduled or planned source reallocation based on a long-term (e.g., one to several days) forecast of anticipated traffic.
This approach is also often inadequate as it relies on the accuracy of a long-term forecast (which may, e.g., underestimate the success of a sales promotion) and is also exposed to unanticipated events (e.g., a traffic surge at news web sites such as experienced at CNN's web site on 9/11/01).
Another key disadvantage of existing computing deployment approaches is that they generally require some form of manual intervention, e.g., via expert operators, to adjust for resource imbalance.
Accordingly, it would be desirable to have automated or autonomic techniques for managing a computing deployment, associated with a computing system or network, which handle variable workload more efficiently and effectively than existing approaches.
Summary of the Invention The present invention provides automated or autonomic techniques for managing a computing deployment, associated with a computing system or network, which handle variable workload more efficiently and effectively than existing approaches.
In one aspect of the invention, techniques are provided for managing deployment of one or more resources in a computing environment based on varying workload levels.
The techniques may comprise predicting a future workload level based on data associated with the computing environment. Then, an estimation is performed to determine whether a current resource deployment is insufficient, sufficient, or overly sufficient to satisfy the future workload level. Then, one or more actions are caused to be taken when the current resource deployment is estimated to be insufficient or overly sufficient to satisfy the future workload level. Actions may comprise resource provisioning, resource tuning and/or admission control Advantageously, the present invention may provide for proactively maintaining a service level objective, such as response time, for a computing deployment in the face of variable workload. In particular, by making changes to a computing deployment in an automated or autonomic fashion, the techniques employed by the invention are effective at accommodating unanticipated workload.
The present invention also advantageously provides a methodology for an application owner to attempt to ensure satisfaction of one or more service objectives associated with the execution of an application that is hosted by a service provider. This may be accomplished by the application owner contracting with the service provider to host the application and to implement a computing deployment management system as provided herein.
Accordingly, in one aspect there is provided an automated method of managing deployment of a plurality of servers in a computing environment based on varying workload levels, the method comprising the steps of:
predicting a future workload level based on data associated with the computing environment;
wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level, by a processor in response to instructions stored on a non-transitory computer readable medium, as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to another aspect there is provided an apparatus for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the apparatus comprising:
a memory; and at least one processor coupled to the memory and operative to:
(i) predict a future workload level based on data associated with the computing environment using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
(iii) cause one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and (iv) selectively adapt the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing operation further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing operation further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to yet another aspect there is provided an article of manufacture for managing deployment of a plurality of servers in a computing environment based on varying workload levels, comprising a computer readable storage medium containing one or more programs which when executed implement the steps of predicting a future workload level based on data associated with the computing environment, wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
3a estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
According to still yet another aspect there is provided an automated system for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the system comprising:
a solution manager module comprising memory and at least one processor coupled thereto and operative to:
(i) predict a future workload level based on data associated with the computing environment, wherein the prediction operation further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level; and (iii) selectively adapt the forecast horizon as a function of a time needed to effectuate at least one of one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and a deployment manager coupled to the solution manager module, comprising a memory and at least one processor coupled thereto and operative to:
(i) provide the data associated with the computing environment to the solution manager module; and 3b (ii) effect the one or more actions to be taken, in response to the solution manager module, when the current deployment of servers is estimated by the solution manager module to be one of insufficient and overly sufficient to satisfy the future workload level, wherein the one or more actions to be taken comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient and wherein the one or more actions to be taken comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient;
wherein the deployment manager further comprises:
(i) a monitoring module for providing access to workload data;
(ii) a provisioning module for performing resource provisioning;
(iii) a tuning interface module for changing one or more configuration parameters associated with one or more of the servers; and (iv) a throttling interface module for causing a manipulation of one or more admission queues on one or more of the servers.
These and other objects, features and advantages of the present invention will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in connection with the accompanying drawings.
Brief Description of the Drawings FIG. 1 is a block diagram illustrating a computing deployment management system according to an embodiment of the present invention and an overall environment in which such system may operate;
FIG. 2 is a flow diagram illustrating a computing deployment management methodology according to an embodiment of the present invention;
FIG. 3 is a graphical representation illustrating performance of a computing system or network in accordance with principles of the present invention; and FIG. 4 is a block diagram illustrating a generalized hardware architecture of a computer system suitable for implementing a computing deployment management system according to the present invention.
3c Detailed Description of Preferred Embodiments The present invention will be explained below in the context of an illustrative web-based computing network environment. That is, the computing resources being managed (e.g., application servers, database connections, input/output paths, etc.) are associated with one or more web sites. However, it is to be understood that the present invention is not limited to such a particular environment. Rather, the invention is more generally applicable to any computing environment in which it is desirable to automatically or autonomically manage and compute resource deployment in the face of variable workload.
As is known, "autonomic" computing generally refers to a comprehensive and holistic approach to self-managed computing systems with a minimum of human interference, see, e.g., P. Horn, "Autonomic Computing IBM's Perspective on the State of Information Technology," IBM Research, October 2001. The teiin derives from the body's autonomic nervous system, which controls key functions without conscious awareness or involvement.
More specifically, one of the goals of autonomic computing is to automate some or all of the tasks an expert operator or administrator would typically carry out. Thus, as will be appreciated from the inventive principles presented herein, the computing deployment techniques of the invention are able to operate automatically or autonomically.
Referring initially to FIG. 1, a block diagram illustrates a computing deployment management system according to an embodiment of the present invention and an overall environment in which such system may operate. As shown, the environment 100 comprises a computing deployment management system 100. The computing deployment management system 100, itself, comprises a solution manager 120 and a deployment manager 130. The solution manager 120, itself, comprises a control logic engine 122, a forecaster module 124, a performance estimator module 126, an admission control module 128, and a tuning module 129. The deployment manager 130, itself comprises a monitoring module 132, a provisioning module 134, a tuning interface module 136, and a throttling interface module 138.
Further, as shown, the environment 100 comprises an application level 140. The application level, itself, comprises resource pool 142 (resource pool A
comprising, for example, application servers), resource pool 144 (resource pool B comprising, for example, database connections), and resource pool 146 (resource pool C
comprising, for example, input/output paths).
Accordingly, the architecture shown in FIG. 1 is organized into three levels:
(a) the application level (denoted as 140) and associated resources on which the application can be deployed; (b) a deployment management level (denoted as 130) which provides connection and control of resources; and (c) a solution management level (denoted as 120) which performs the real-time analysis and planning required to initiate actions that are required to maintain a service level objective. These three levels are discussed in further detail below.
Application deployment typically requires a mix of resources of various types, such as, for example, an HTTP (hypertext transport protocol) server, an application server, a database server, storage, connections, I/O paths, etc. In a typical computing deployment (e.g., a data center), these resources could be available from a managed pool.
FIG. 1 illustrates three such managed resource pools 142, 144 and 146. It is understood that, depending on the application, a predetermined number of each of the resources, sufficient to satisfy anticipated workloads, is available for use in the managed resource pools. It is to be further understood that while FIG. 1, and the above description, mention certain resources, the invention is not limited to any particular resources. Rather, the invention may manage any and all types of resources including, but not limited to, hardware components, software components, and combinations thereof It is to be understood that a resource may also be an application, itself, or some portion thereof.
The deployment manager 130 interfaces with relevant resources of the application level 140 to monitor measurement/configuration data (e.g., through resource-dependent sensors such as, for example, response time probes, vmstat data from the operating system such as Unix, snapshot data from a database such as IBM Corporation's DB2, or through custom interfaces or standard interfaces implemented using the Common Information Model) and to control the resources (e.g., through resource-dependent effectuators such as, for example, the node agent on an application server such as IBM
Corporation's WebSphere Application Server, the application programming interface for changing configuration parameters in a database such as IBM Corporation's DB2). Hence, the deployment manager is able to perform resource provisioning (via provisioning module 134) which, by way of example, for a piece of hardware, can range from: (i) deploying an operating system on a computer without an installed operating system, e.g., an x86 system on which Windows or Linux can be installed, or replacing an existing operating system on a computer with a new operating system; (ii) deploying appropriate middleware on top of the operating system; (iii) deploying an application with associated data; and (iv) performing relevant cluster management/federation to enable an added resource to support the application. Advantageous features of this provisioning capability include not only rapidly and automatically adding resources when needed, for example, in response to an unexpected workload surge, but also removing resources when no longer needed, hence minimizing the greater cost of additional resources.
In addition, the deployment manager 130 (via tuning interface module 136) resets resource configuration parameters (e.g., memory pool sizes such as buffer pools for a database, ORB (object request broker) thread pool size in an application server) which is important for resource tuning. Resource tuning generally refers to the technique of changing one or more configuration parameters associated with a resource in a manner which helps achieve a goal such as minimizing response time or maximizing throughput.
The deployment manager 130 (via throttling interface module 138) also manipulates admission queues on the resources (e.g., for admission control/request throttling).
Throttling generally refers to rejecting incoming requests based on some policies that identify service classes such as, for example, type of request (buy versus browse at an e-commerce site which can be distinguished by a uniform resource locator), origin (preferred customers), etc. By rejecting such requests, the incoming load to a computing deployment may be reduced to a manageable level. The deployment manager (via monitoring module 132) also provides access to the workload data (e.g., throughput, response time, etc.).
The solution manager 120 is responsible for maintaining the service objective for the particular application deployment. "Service objective" may refer to requirements and/or preferences specified in accordance with a service level agreement (SLA). That is, by way of example, such service objectives may deal with how service applications are hosted at a third party infrastructure, while ensuring a certain level of end-client satisfaction. As is known, businesses increasingly run their applications using infrastructure (e.g., server, network connectivity) provided by a third party, generally referred to as the "service provider." Many companies, such as IBM Global Services, host web sites and/or provide other computer hosting services. An SLA provides a means by which the expectations of the service provider can be negotiated with the customer. An SLA between an application owner and the service provider defines terms and conditions for this hosting service. The SLA may, for example, include expected response time, bandwidth throughput at the network and/or servers, disk space utilization, availability, i.e., up-time of network and server resources, as well recovery time upon failure, and pricing for various levels of service. However, it is to be appreciated that a service objective does not have to come from an SLA, which typically has legal consequences. A
service level objective can often be negotiated within an enterprise, e.g., between the information technology (IT) department and the purchasing department for whom they may be deploying an online purchase order system. Also an e-commerce site or even a , , place like Google may want to maintain a good service level, with regard to something like response time, so that the user experience is good.
Accordingly, so as to sufficiently maintain the particular service objectives, the solution manager 120, in accordance with the control logic engine 122, decides: (i) when action needs to be taken; and (ii) what action to take. The control logic engine 122 accomplishes these tasks, as will be explained below, in accordance with forecaster module 124, performance estimator module 126, admission control module 128, and tuning module 129. That is, it is to be understood that the control logic engine (CLE) 122 serves as a controller for the functions provided by the other modules in the solution manager 120. It is to be understood, however, that the functional arrangement shown in block 120 is illustrative in nature and, thus, other arrangements for controlling the functionality provided by the solution manager may be employed within the scope of the principles of the present invention.
One advantageous feature of the solution manager is the ability to proactively manage the service objective, hence, minimizing/avoiding a possible violation.
This is accomplished by a combination of forecasting (via forecaster module 124) on the workload (obtained from the monitoring component in the deployment manager) in combination with performance estimation (via performance estimator 126) to check if a service level violation is anticipated. In a preferred embodiment, the control logic component 122 is essentially a rule based component that, based on the forecasting and performance estimation results, decides which action or actions to take.
Examples of actions that can be taken include provisioning (adding/removing resources), admission control, and resource tuning.
Resource tuning may be effective when the workload changes. For instance, in a e-commerce site, the relative buy to browse mix can change, and since these transactions can draw on somewhat different resources in a database, e.g., such as use different buffer pools, or require more sorting, the overall system responsiveness may be improved by changing database configuration parameters, such as relative buffer pool sizes, or sort heaps. This is determined and accomplished by the tuning module 129, via the tuning interface 136.
Admission control actually rejects some of the incoming requests based on some policy, e.g., admit only preferred customers, or only buy transactions, such that the system is not overloaded. This kind of action may be taken if others actions such as provisioning or resource tuning cannot achieve the desired results.
In any case, the magnitude of the action (e.g., how many servers to add, how much to change a configuration parameter) is determined using the performance estimation capability. These actions could be taken separately or in combination.
The following is an illustrative description of a functional operation of a computing deployment management system according to the present invention. In this example, one type of control action is considered, namely, application provisioning, which is accomplished by server addition to/removal from the active cluster.
Referring now to FIG. 2, a flow diagram illustrates a computing deployment management methodology 200 according to an embodiment of the present invention.
The solution manager 120 keeps track of the workload (transaction rate), service objective (response time), and configuration (resources which are active, idle, and in transition states) that the control logic engine (CLE) 122 obtains from the monitoring module 132 in the deployment manager 130 (step 202).
Based on monitored data such as the transaction rate history, the solution manager 120 uses the forecaster module 124 to project into the future the expected workload (step 204).
The forecast can be long-term (e.g., hours/days) or short-term (e.g., seconds/minutes). To deal with unanticipated changes in the workload (e.g., outside of normal daily/weekly, etc.) variations, the forecaster is preferably adaptive and rapidly learns from recent history. The forecast horizon does not need to be long, e.g., the horizon may be approximately the amount of time needed for the resource action (application provisioning in this situation) to take effect.
In a preferred embodiment, the forecaster module 124 may be implemented using the techniques described in U.S. Patent 7,039,559 entitled "Methods and Apparatus for Performing Adaptive and Robust Prediction," filed concurrently herewith.
However, it is to be appreciated that the forecaster module 124 may employ other forecasting techniques such as, for example, those described in "Time Series Analysis: Forecasting and Control,"
by G.E.P. Box et al. (rev. ed.), or those employed in the Hyperion Forecasting Suite available from Hyperion.
The solution manager 120 then uses the performance estimator 126 to check if the current resource deployment is insufficient, sufficient (i.e., adequate) or overly sufficient (i.e., excessive) for maintaining the service objective threshold (e.g., response time) based on the recent and forecasted workload traffic (step 206). While the invention is not limited to any particular performance estimation techniques, examples of such techniques that may be employed include those described in "Configuration and Capacity Planning for Solaris Servers," by Brian L Wong, and "Capacity Planning for Web Services:
Metrics, Models, and Methods" by Daniel A. Menasce et al., and/or those referred to as PATROL Perfom and PATROL Predict from BMC Software, and High Volume Web Site Performance Simulator for Websphere from IBM Corporation. Based on recommendations of the performance estimator, and knowledge of state and number servers in the relevant resource pool, the solution manager 120 sends a request to the application provisioner 134 in the deployment manager 130 to add/remove the appropriate number of servers, if necessary (step 208). That is, if it is determined by the CLE 122 that the current resources are sufficient, then no provisioning may be necessary at this time.
, If provisioning is needed (step 210), the provisioner 134 acts rapidly to add a server from the idle state to the active cluster running the application.
Rapid addition (e.g., from about several minutes to less than about one minute) may be achieved by a combination of loading saved application images on the server instance being added, starting the server, and allowing the cluster manager to send requests to it.
Note that speed of resource addition may be increased if the servers are preloaded with the application image. Hence, the preloaded servers need only to be started and activated within the cluster. To remove the server from the active cluster, the provisioner 134 disables new incoming requests to the server from the cluster manager, while allowing existing work to continue. This allows a natural quiescing of the work on that server, which is then stopped.
It is to be appreciated that by use herein of the phrases "insufficient" and "overly sufficient" (and, similarly, "inadequate" and "excessive") with respect to computing resources, it is generally not only meant that there are not enough (insufficient) resources deployed or too many (overly sufficient) resources deployed, but also that one or more resources are improperly configured to meet the anticipated workload (e.g., due to a workload mix change, etc.) or that certain admission requests that should not be admitted are being admitted Thus, in accordance with the invention, the resource deployment may be made "sufficient" (or, similarly, "adequate") not only by adding or removing resources, but also by tuning one or more resources and/or throttling certain admission requests.
The following description provides some results from an exemplary implementation of the management system described herein. In the exemplary deployment considered, the application is a supply chain management. The application is run on a cluster of web application servers with a database server in the back-end. In this example, the system is faced with a large unanticipated surge of incoming transactions requests (doubling every minute to about 20 times the normal load) while trying to stay below a response time service objective of two seconds. The prediction horizon of the short term forecaster is one minute, since the rapid application deployment capability of the provisioner is about 30 to 40 seconds.
The performance of the system is shown in FIG. 3. The prediction of the transaction rate (top panel of FIG. 3, line 302) initially lags the surge in the actual transaction rate (top panel of FIG. 3, line 304), but very quickly catches up and is able to provide useful guidance on the anticipated transaction rate. This is in turn used to calculate the requisite number of application servers using the performance estimator (126 of FIG. 1), and the servers are quickly brought into active (middle panel of FIG. 3, line 306) duty with the rapid provisioning capability of the application provisioner (134 of FIG. 1). The amount of time a server is in the transitional starting state (middle panel of FIG. 3, line 308) while going from idle to active is only about 30 seconds.
The combination of the forecasting and performance estimation in the solution manager together with the rapid application provisioning capability of the deployment manager allows the system to stay below the two second response time target (bottom panel of FIG. 3, line 310) in the face of an extremely aggressive surge. After the surge is over, the servers are returned to the pool.
Referring now to FIG. 4, a block diagram illustrates a generalized hardware architecture of a computer system suitable for implementing a computing deployment management system according to the present invention. For instance, the functional components shown in FIG. 1 with respect to the solution manager 120 and the deployment manager 130 may be implemented on one or more computer systems of the type shown in FIG. 4. Of course, separate functional components may be implemented on their own dedicated computer system. However, it is to be appreciated that the computing deployment management system of the invention is not intended to be limited to any particular computer platform, arrangement or implementation.
In this illustrative implementation 400, a processor 402 for implementing management methodologies and functionalities of the invention as described herein is operatively coupled to a memory 404 and input/output (I/0) devices 406, via a bus 408 or an alternative connection arrangement. It is to be appreciated that the term "processor" as used herein is intended to include any processing device, such as, for example, one that includes a central processing unit (CPU) and/or other processing circuitry (e.g., digital signal processor (DSP), microprocessor, etc.). Additionally, it is to be understood that the term "processor" may refer to more than one processing device, and that various elements associated with a processing device may be shared by other processing devices.
The term "memory" as used herein is intended to include memory and other computer-readable media associated with a processor or CPU, such as, for example, random access memory (RAM), read only memory (ROM), fixed storage media (e.g., hard drive), removable storage media (e.g., diskette), flash memory, etc. The memory may preferably be used to store data and computer programs associated with the invention.
In addition, the term "I/O devices" as used herein is intended to include one or more input devices (e.g., keyboard, mouse) for inputting data to the processing unit, as well as one or more output devices (e.g., CRT display) for providing results associated with the processing unit.
It is to be appreciated that the methodologies of the present invention are capable of being implemented in the form of computer readable media. The term "computer readable media" as used herein is intended to include recordable-type media, such as, for example, a floppy disk, a hard disk drive, RAM, compact disk (CD) ROM, etc., as well as transmission-type media.
Accordingly, one or more computer programs, or software components thereof, including instructions or code for performing the methodologies of the invention, as described herein, may be stored in one or more of the associated storage media (e.g., ROM, fixed or removable storage) and, when ready to be utilized, loaded in whole or in part (e.g., into RAM) and executed by the processor 402.
In any case, it is to be appreciated that the techniques of the invention, described herein and shown in the appended figures, may be implemented in various forms of hardware, software, or combinations thereof, e.g., one or more operatively programmed general purpose digital computers with associated memory, implementation-specific integrated circuit(s), functional circuitry, etc. Given the techniques of the invention provided herein, one of ordinary skill in the art will be able to contemplate other implementations of the techniques of the invention.
Accordingly, as explained herein in detail, the present invention provides an architecture which enables automated proactive management of a system in the face of variable workload or unexpected workload variability. That is, a system for proactive management of computer applications that experience unexpected variability in demand is provided which may comprise components for forecasting, performance modeling, control, and reconfiguration. The forecasting is used to anticipate future workload and actions are taken sufficiently rapidly to accommodate the anticipated workload, while reducing exposure to potential violations of a service level objective.
Actions are taken in a cost-effective manner that minimizes resource under-utilization while avoiding excessive repeated resource configuration changes, i.e., oscillations. Actions taken are based on models/estimations of performance based on available resources together with anticipated workload and consideration of the service objective. Actions may include resource provisioning and/or resource tuning and/or admission control.
An architecture of the invention may be structured into: (a) an application to be managed; (b) a deployment manager that provides a generic interface to that system for sensing and effecting control; and (c) a solution manager that determines when actions are to be taken based on this generic interface to monitoring, and requests the necessary actions through the generic control interface. The deployment manager can manage multiple configurations, allowing for rapid deployment or redeployment of resources as a result of being able to: (a) take actions specific to a particular resource node, a particular configuration, or particular function; and (b) automatically set application appropriate database paths and parameters.
As is further evident, the invention further provides techniques operating an adaptive system in which: (a) workload and service level metrics are collected; (b) future values of workload measurements are forecast; (c) these forecasted values and the desired service level objective are used to determine the actions required to ensure adequate service levels; and (d) the actions themselves are initiated in a way that minimizes oscillations and yet responds promptly.
Although illustrative embodiments of the present invention have been described herein with reference to the accompanying drawings, it is to be understood that the invention is not limited to those precise embodiments, and that various other changes and modifications may be made therein by one skilled in the art without departing from the scope of the invention.
Claims (28)
1. An automated method of managing deployment of a plurality of servers in a computing environment based on varying workload levels, the method comprising the steps of:
predicting a future workload level based on data associated with the computing environment;
wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level, by a processor in response to instructions stored on a non-transitory computer readable medium, as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
predicting a future workload level based on data associated with the computing environment;
wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level, by a processor in response to instructions stored on a non-transitory computer readable medium, as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
2. The method of claim 1, further comprising the step of obtaining the data associated with the computing environment, used by the future workload level predicting step, via monitoring one or more of the servers.
3. The method of claim 1, wherein the action causing step further comprises causing the tuning of one or more configuration parameters associated with the servers.
4. The method of claim 1, wherein the action causing step further comprises causing the manipulation of admission queues on the servers.
5. The method of claim 1, wherein the estimating step further comprises estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient based on one or more service objectives.
6. The method of claim 1, wherein the action causing step further comprises deploying an operating system on a computer without an installed operating system or replacing an existing operating system.
7. The method of claim 1, wherein the action causing step further comprises deploying middleware on top of an operating system.
8. The method of claim 1, wherein the action causing step further comprises deploying an application with associated data.
9. The method of claim 1, wherein the action causing step further comprises performing cluster management to enable an added server to support an application.
10. An apparatus for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the apparatus comprising:
a memory; and at least one processor coupled to the memory and operative to:
(i) predict a future workload level based on data associated with the computing environment using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
(iii) cause one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and (iv) selectively adapt the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing operation further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing operation further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
a memory; and at least one processor coupled to the memory and operative to:
(i) predict a future workload level based on data associated with the computing environment using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
(iii) cause one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and (iv) selectively adapt the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing operation further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing operation further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
11. The apparatus of claim 10, wherein the at least one processor is further operative to obtain the data associated with the computing environment, used by the future workload level predicting operation, via monitoring one or more of the servers.
12. The apparatus of claim 10, wherein the action causing operation further comprises causing the tuning of one or more configuration parameters associated with the servers.
13. The apparatus of claim 10, wherein the action causing operation further comprises causing the manipulation of admission queues on the servers.
14. The apparatus of claim 10, wherein the estimating operation further comprises estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient based on one or more service objectives.
15. An article of manufacture for managing deployment of a plurality of servers in a computing environment based on varying workload levels, comprising a computer readable storage medium including a recordable-type medium containing one or more programs which when executed by the computer implement the steps of:
predicting a future workload level based on data associated with the computing environment, wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
predicting a future workload level based on data associated with the computing environment, wherein the predicting step further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level;
causing one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and selectively adapting the forecast horizon used to calculate the future workload level as a function of a time needed to effectuate at least one of the one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level;
wherein the action causing step further comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient; and wherein the action causing step further comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient.
16. The article of claim 15, further comprising the step of obtaining the data associated with the computing environment, used by the future workload level predicting step, via monitoring one or more of the servers.
17. The article of claim 15, wherein the action causing step further comprises causing the tuning of one or more configuration parameters associated with the servers.
18. The article of claim 15, wherein the action causing step further comprises causing the manipulation of admission queues on the servers.
19. The article of claim 15, wherein the estimating step further comprises estimating whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient based on one or more service objectives.
20. An automated system for managing deployment of a plurality of servers in a computing environment based on varying workload levels, the system comprising:
a solution manager module comprising memory and at least one processor coupled thereto and operative to:
(i) predict a future workload level based on data associated with the computing environment, wherein the prediction operation further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level; and (iii) selectively adapt the forecast horizon as a function of a time needed to effectuate at least one of one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and a deployment manager coupled to the solution manager module, comprising a memory and at least one processor coupled thereto and operative to:
(i) provide the data associated with the computing environment to the solution manager module; and (ii) effect the one or more actions to be taken, in response to the solution manager module, when the current deployment of servers is estimated by the solution manager module to be one of insufficient and overly sufficient to satisfy the future workload level, wherein the one or more actions to be taken comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient, and wherein the one or more actions to be taken comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient;
wherein the deployment manager further comprises:
(i) a monitoring module for providing access to workload data;
(ii) a provisioning module for performing resource provisioning;
(iii) a tuning interface module for changing one or more configuration parameters associated with one or more of the servers; and (iv) a throttling interface module for causing a manipulation of one or more admission queues on one or more of the servers.
a solution manager module comprising memory and at least one processor coupled thereto and operative to:
(i) predict a future workload level based on data associated with the computing environment, wherein the prediction operation further comprises forecasting using a forecasting equation based on a forecast horizon, the forecasting equation using the forecast horizon to calculate the future workload level;
(ii) estimate whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient to satisfy the future workload level; and (iii) selectively adapt the forecast horizon as a function of a time needed to effectuate at least one of one or more actions to be taken when the current deployment of servers is estimated to be one of insufficient and overly sufficient to satisfy the future workload level; and a deployment manager coupled to the solution manager module, comprising a memory and at least one processor coupled thereto and operative to:
(i) provide the data associated with the computing environment to the solution manager module; and (ii) effect the one or more actions to be taken, in response to the solution manager module, when the current deployment of servers is estimated by the solution manager module to be one of insufficient and overly sufficient to satisfy the future workload level, wherein the one or more actions to be taken comprises causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient, and wherein the one or more actions to be taken comprises causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient;
wherein the deployment manager further comprises:
(i) a monitoring module for providing access to workload data;
(ii) a provisioning module for performing resource provisioning;
(iii) a tuning interface module for changing one or more configuration parameters associated with one or more of the servers; and (iv) a throttling interface module for causing a manipulation of one or more admission queues on one or more of the servers.
21. The system of claim 20, wherein the servers are deployable to implement execution of an application.
22. The system of claim 20, wherein at least one of the solution manager and deployment manager operate autonomically.
23. The system of claim 20, wherein the one or more actions comprise at least one of server provisioning, server tuning, and admission control.
24. The system of claim 20, wherein the solution manager estimates whether a current deployment of servers is one of insufficient, sufficient, and overly sufficient based on one or more service objectives.
25. The method of any one of claims 1 to 9, wherein causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient by:
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
26. The apparatus of any one of claims 10 to 14, wherein causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient by:
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
27. The article of any one of claims 15 to 19, wherein causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient by:
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
adding a server instance loaded with saved application images associated with an application running in the computing environment; and allowing requests to be sent to the server instance; and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by:
disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
28. The automated system of any one of claims 20 to 24, wherein causing the addition of one or more servers to address the future workload level when the current deployment of servers is estimated to be insufficient by adding a server instance loaded with saved application images associated with an application running in the computing environment and allowing requests to be sent to the server instance, and wherein causing the removal of one or more servers to address the future workload level when the current deployment of servers is estimated to be overly sufficient by disabling new incoming requests to a server instance; and allowing existing work on the server instance to continue until completed.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/384,973 US7350186B2 (en) | 2003-03-10 | 2003-03-10 | Methods and apparatus for managing computing deployment in presence of variable workload |
US10/384,973 | 2003-03-10 | ||
PCT/US2003/027304 WO2004081789A2 (en) | 2003-03-10 | 2003-08-29 | Methods and apparatus for managing computing deployment in presence of variable workload |
Publications (2)
Publication Number | Publication Date |
---|---|
CA2515470A1 CA2515470A1 (en) | 2004-09-23 |
CA2515470C true CA2515470C (en) | 2016-10-25 |
Family
ID=32961408
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA2515470A Expired - Lifetime CA2515470C (en) | 2003-03-10 | 2003-08-29 | Methods and apparatus for managing computing deployment in presence of variable workload |
Country Status (10)
Country | Link |
---|---|
US (2) | US7350186B2 (en) |
EP (1) | EP1602031A2 (en) |
JP (1) | JP2006520027A (en) |
KR (1) | KR100826833B1 (en) |
CN (1) | CN100377094C (en) |
AU (1) | AU2003268329A1 (en) |
CA (1) | CA2515470C (en) |
MX (1) | MXPA05009648A (en) |
TW (1) | TWI308269B (en) |
WO (1) | WO2004081789A2 (en) |
Families Citing this family (185)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6907395B1 (en) * | 2000-10-24 | 2005-06-14 | Microsoft Corporation | System and method for designing a logical model of a distributed computer system and deploying physical resources according to the logical model |
US7606898B1 (en) * | 2000-10-24 | 2009-10-20 | Microsoft Corporation | System and method for distributed management of shared computers |
GB0220846D0 (en) * | 2002-09-07 | 2002-10-16 | Ibm | Remote dynamic configuration of a web server to facilitate capacity on demand |
US7472272B2 (en) * | 2003-01-23 | 2008-12-30 | Verdasys, Inc. | Digital asset usage accountability via event journaling |
US7765501B2 (en) * | 2003-03-06 | 2010-07-27 | Microsoft Corporation | Settings and constraints validation to enable design for operations |
US7689676B2 (en) * | 2003-03-06 | 2010-03-30 | Microsoft Corporation | Model-based policy application |
US7890543B2 (en) | 2003-03-06 | 2011-02-15 | Microsoft Corporation | Architecture for distributed computing system and automated design, deployment, and management of distributed applications |
US8122106B2 (en) | 2003-03-06 | 2012-02-21 | Microsoft Corporation | Integrating design, deployment, and management phases for systems |
US7350186B2 (en) * | 2003-03-10 | 2008-03-25 | International Business Machines Corporation | Methods and apparatus for managing computing deployment in presence of variable workload |
US7039559B2 (en) | 2003-03-10 | 2006-05-02 | International Business Machines Corporation | Methods and apparatus for performing adaptive and robust prediction |
US7814126B2 (en) * | 2003-06-25 | 2010-10-12 | Microsoft Corporation | Using task sequences to manage devices |
US7590736B2 (en) * | 2003-06-30 | 2009-09-15 | Microsoft Corporation | Flexible network load balancing |
US7606929B2 (en) * | 2003-06-30 | 2009-10-20 | Microsoft Corporation | Network load balancing with connection manipulation |
US7636917B2 (en) * | 2003-06-30 | 2009-12-22 | Microsoft Corporation | Network load balancing with host status information |
US7103874B2 (en) * | 2003-10-23 | 2006-09-05 | Microsoft Corporation | Model-based management of computer systems and distributed applications |
JP2005141441A (en) * | 2003-11-06 | 2005-06-02 | Hitachi Ltd | Load distribution system |
US7827535B2 (en) * | 2003-12-10 | 2010-11-02 | Oracle International Corporation | Application performance tuning server-side component |
US7757216B2 (en) * | 2003-12-10 | 2010-07-13 | Orcle International Corporation | Application server performance tuning client interface |
US7734561B2 (en) * | 2003-12-15 | 2010-06-08 | International Business Machines Corporation | System and method for providing autonomic management of a networked system using an action-centric approach |
US8145731B2 (en) * | 2003-12-17 | 2012-03-27 | Hewlett-Packard Development Company, L.P. | System and method for determining how many servers of at least one server configuration to be included at a service provider's site for supporting an expected workload |
US7778422B2 (en) | 2004-02-27 | 2010-08-17 | Microsoft Corporation | Security associations for devices |
US8782654B2 (en) | 2004-03-13 | 2014-07-15 | Adaptive Computing Enterprises, Inc. | Co-allocating a reservation spanning different compute resources types |
US20050246529A1 (en) * | 2004-04-30 | 2005-11-03 | Microsoft Corporation | Isolated persistent identity storage for authentication of computing devies |
US20070266388A1 (en) | 2004-06-18 | 2007-11-15 | Cluster Resources, Inc. | System and method for providing advanced reservations in a compute environment |
US7912940B2 (en) * | 2004-07-30 | 2011-03-22 | Microsoft Corporation | Network system role determination |
US20060026503A1 (en) * | 2004-07-30 | 2006-02-02 | Wireless Services Corporation | Markup document appearance manager |
US8176490B1 (en) | 2004-08-20 | 2012-05-08 | Adaptive Computing Enterprises, Inc. | System and method of interfacing a workload manager and scheduler with an identity manager |
JP2006072785A (en) * | 2004-09-03 | 2006-03-16 | Hitachi Electronics Service Co Ltd | Request message control method for using of service, and service provision system |
US20060070060A1 (en) * | 2004-09-28 | 2006-03-30 | International Business Machines Corporation | Coordinating service performance and application placement management |
JP4167643B2 (en) * | 2004-10-27 | 2008-10-15 | 株式会社日立製作所 | Business system operation method, operation management system, and operation program |
CA2586763C (en) | 2004-11-08 | 2013-12-17 | Cluster Resources, Inc. | System and method of providing system jobs within a compute environment |
US7693982B2 (en) * | 2004-11-12 | 2010-04-06 | Hewlett-Packard Development Company, L.P. | Automated diagnosis and forecasting of service level objective states |
US7496575B2 (en) * | 2004-11-22 | 2009-02-24 | Verdasys, Inc. | Application instrumentation and monitoring |
US20060112155A1 (en) * | 2004-11-24 | 2006-05-25 | Agami Systems, Inc. | System and method for managing quality of service for a storage system |
US7836451B2 (en) * | 2004-12-14 | 2010-11-16 | International Business Machines Corporation | Method, system and program product for approximating resource consumption of a computer system |
US20060130042A1 (en) * | 2004-12-15 | 2006-06-15 | Dias Daniel M | Method and apparatus for dynamic application upgrade in cluster and grid systems for supporting service level agreements |
US8316130B2 (en) | 2004-12-22 | 2012-11-20 | International Business Machines Corporation | System, method and computer program product for provisioning of resources and service environments |
US7516206B2 (en) * | 2005-01-28 | 2009-04-07 | Cassatt Corporation | Management of software images for computing nodes of a distributed computing system |
US8387037B2 (en) * | 2005-01-28 | 2013-02-26 | Ca, Inc. | Updating software images associated with a distributed computing system |
US7680799B2 (en) * | 2005-01-31 | 2010-03-16 | Computer Associates Think, Inc. | Autonomic control of a distributed computing system in accordance with a hierarchical model |
US7454427B2 (en) * | 2005-01-31 | 2008-11-18 | Cassatt Corporation | Autonomic control of a distributed computing system using rule-based sensor definitions |
US7685148B2 (en) * | 2005-01-31 | 2010-03-23 | Computer Associates Think, Inc. | Automatically configuring a distributed computing system according to a hierarchical model |
US7478097B2 (en) | 2005-01-31 | 2009-01-13 | Cassatt Corporation | Application governor providing application-level autonomic control within a distributed computing system |
US7571154B2 (en) | 2005-01-31 | 2009-08-04 | Cassatt Corporation | Autonomic control of a distributed computing system using an application matrix to control application deployment |
US7590653B2 (en) * | 2005-03-02 | 2009-09-15 | Cassatt Corporation | Automated discovery and inventory of nodes within an autonomic distributed computing system |
US8863143B2 (en) | 2006-03-16 | 2014-10-14 | Adaptive Computing Enterprises, Inc. | System and method for managing a hybrid compute environment |
US9075657B2 (en) | 2005-04-07 | 2015-07-07 | Adaptive Computing Enterprises, Inc. | On-demand access to compute resources |
US9015324B2 (en) | 2005-03-16 | 2015-04-21 | Adaptive Computing Enterprises, Inc. | System and method of brokering cloud computing resources |
CA2601384A1 (en) | 2005-03-16 | 2006-10-26 | Cluster Resources, Inc. | Automatic workload transfer to an on-demand center |
US9231886B2 (en) | 2005-03-16 | 2016-01-05 | Adaptive Computing Enterprises, Inc. | Simple integration of an on-demand compute environment |
JP4616674B2 (en) * | 2005-03-18 | 2011-01-19 | 株式会社日立製作所 | Resource lending method and resource lending system |
US7908314B2 (en) * | 2005-03-23 | 2011-03-15 | Hitachi, Ltd. | Method for controlling a management computer |
US8782120B2 (en) | 2005-04-07 | 2014-07-15 | Adaptive Computing Enterprises, Inc. | Elastic management of compute resources between a web server and an on-demand compute environment |
US7383161B2 (en) * | 2005-04-13 | 2008-06-03 | Microsoft Corporation | Systems and methods for device simulation |
US7383516B2 (en) * | 2005-04-13 | 2008-06-03 | Microsoft Corporation | Systems and methods for displaying and editing hierarchical data |
US8489728B2 (en) | 2005-04-15 | 2013-07-16 | Microsoft Corporation | Model-based system monitoring |
US7552036B2 (en) | 2005-04-15 | 2009-06-23 | Microsoft Corporation | Preconditioning for stochastic simulation of computer system performance |
US7797147B2 (en) | 2005-04-15 | 2010-09-14 | Microsoft Corporation | Model-based system monitoring |
US7979520B2 (en) * | 2005-04-15 | 2011-07-12 | Microsoft Corporation | Prescriptive architecture recommendations |
US7689616B2 (en) * | 2005-04-15 | 2010-03-30 | Microsoft Corporation | Techniques for specifying and collecting data aggregations |
US20060242647A1 (en) * | 2005-04-21 | 2006-10-26 | Kimbrel Tracy J | Dynamic application placement under service and memory constraints |
US7979859B2 (en) * | 2005-05-03 | 2011-07-12 | International Business Machines Corporation | Managing automated resource provisioning with a workload scheduler |
US7831976B2 (en) * | 2005-05-04 | 2010-11-09 | International Business Machines Corporation | Method, system and program product for predicting computer system resource consumption |
US7693983B1 (en) * | 2005-05-27 | 2010-04-06 | Symantec Operating Corporation | System and method providing application redeployment mappings using filtered resource usage data |
US7392159B2 (en) * | 2005-06-20 | 2008-06-24 | International Business Machines Corporation | Method and apparatus of capacity learning for computer systems and applications |
US20070016393A1 (en) * | 2005-06-29 | 2007-01-18 | Microsoft Corporation | Model-based propagation of attributes |
US8549513B2 (en) * | 2005-06-29 | 2013-10-01 | Microsoft Corporation | Model-based virtual system provisioning |
US9104650B2 (en) | 2005-07-11 | 2015-08-11 | Brooks Automation, Inc. | Intelligent condition monitoring and fault diagnostic system for preventative maintenance |
KR101322434B1 (en) * | 2005-07-11 | 2013-10-28 | 브룩스 오토메이션 인코퍼레이티드 | Intelligent condition-monitoring and fault diagnostic system |
US8175906B2 (en) * | 2005-08-12 | 2012-05-08 | International Business Machines Corporation | Integrating performance, sizing, and provisioning techniques with a business process |
US20070100760A1 (en) * | 2005-10-31 | 2007-05-03 | Caterpillar Inc. | System and method for selling work machine projects |
US7941309B2 (en) | 2005-11-02 | 2011-05-10 | Microsoft Corporation | Modeling IT operations/policies |
US7668175B2 (en) * | 2005-11-22 | 2010-02-23 | Sun Microsystems, Inc. | Dynamic power management for I/O resource pools |
EP1955152A1 (en) * | 2005-12-01 | 2008-08-13 | Cassatt Corporation | Automated deployment and configuration of applications in an autonomically controlled distributed computing system |
WO2007064799A1 (en) * | 2005-12-01 | 2007-06-07 | Cassatt Corporation | Automated deployment and configuration of applications in an autonomically controlled distributed computing system |
US7269599B2 (en) * | 2005-12-01 | 2007-09-11 | International Business Machines Corporation | Method and system for predicting user activity levels associated with an application |
US20070143686A1 (en) * | 2005-12-15 | 2007-06-21 | International Business Machines Corporation | System administration console that integrates manual and autonomic tasks |
US8782641B2 (en) * | 2006-01-24 | 2014-07-15 | International Business Machines Corporation | Tuning of work to meet performance goal |
JP4605036B2 (en) * | 2006-01-27 | 2011-01-05 | 日本電気株式会社 | Computer system, management server, method and program for reducing computer setting time |
US7844441B2 (en) * | 2006-03-27 | 2010-11-30 | International Business Machines Corporation | Computer-implemented method, system and program product for approximating resource consumption of computer system |
US8572138B2 (en) * | 2006-03-30 | 2013-10-29 | Ca, Inc. | Distributed computing system having autonomic deployment of virtual machine disk images |
US8225311B1 (en) * | 2006-03-30 | 2012-07-17 | Emc Corporation | Deploying and distributing content management code |
US9128766B1 (en) * | 2006-04-24 | 2015-09-08 | Hewlett-Packard Development Company, L.P. | Computer workload redistribution schedule |
US7756973B2 (en) * | 2006-04-27 | 2010-07-13 | International Business Machines Corporation | Identifying a configuration for an application in a production environment |
US8266616B1 (en) * | 2006-05-11 | 2012-09-11 | Hewlett-Packard Development Company, L.P. | Computer system provisioning using templates |
US9785477B2 (en) * | 2006-06-05 | 2017-10-10 | International Business Machines Corporation | Providing a policy hierarchy in an enterprise data processing system |
EP2080124A4 (en) * | 2006-07-09 | 2013-10-16 | Microsoft Amalgamated Company Iii | Systems and methods for managing networks |
US8555287B2 (en) | 2006-08-31 | 2013-10-08 | Bmc Software, Inc. | Automated capacity provisioning method using historical performance data |
US20080077932A1 (en) * | 2006-09-25 | 2008-03-27 | International Business Machines Corporation | Mechanism for Automatically Managing the Resource Consumption of Transactional Workloads |
US9251498B2 (en) * | 2006-10-23 | 2016-02-02 | Oracle International Corporation | Facilitating deployment of customizations of enterprise applications |
US7889677B1 (en) * | 2006-10-30 | 2011-02-15 | At&T Mobility Ii Llc | SS7 network planning and forecasting tool |
US20080109390A1 (en) * | 2006-11-03 | 2008-05-08 | Iszlai Gabriel G | Method for dynamically managing a performance model for a data center |
US20080114879A1 (en) * | 2006-11-14 | 2008-05-15 | Microsoft Corporation | Deployment of configuration data within a server farm |
US7987462B2 (en) * | 2006-11-16 | 2011-07-26 | International Business Machines Corporation | Method for automatic throttling of work producers |
US8331229B1 (en) | 2006-12-15 | 2012-12-11 | At&T Mobility Ii Llc | Policy-enabled dynamic deep packet inspection for telecommunications networks |
US20080168310A1 (en) * | 2007-01-05 | 2008-07-10 | Microsoft Corporation | Hardware diagnostics and software recovery on headless server appliances |
US9432443B1 (en) * | 2007-01-31 | 2016-08-30 | Hewlett Packard Enterprise Development Lp | Multi-variate computer resource allocation |
US20140019284A1 (en) * | 2012-07-12 | 2014-01-16 | Wefi, Inc. | Methods, Systems, and Computer-Readable Media for Network Capacity Allocation |
US8046767B2 (en) * | 2007-04-30 | 2011-10-25 | Hewlett-Packard Development Company, L.P. | Systems and methods for providing capacity management of resource pools for servicing workloads |
US8301740B2 (en) * | 2007-06-27 | 2012-10-30 | Ca, Inc. | Autonomic control of a distributed computing system using dynamically assembled resource chains |
US7895317B2 (en) * | 2007-06-27 | 2011-02-22 | Computer Associates Think, Inc. | Autonomic control of a distributed computing system using finite state machines |
US8041773B2 (en) | 2007-09-24 | 2011-10-18 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
US7814206B1 (en) * | 2007-10-05 | 2010-10-12 | At&T Mobility Ii Llc | Forecasting tool for communications network platforms |
KR100972120B1 (en) * | 2008-02-26 | 2010-07-26 | 한국과학기술연구원 | Dynamic robot software architecture management method based on computing resources |
US20090293051A1 (en) * | 2008-05-22 | 2009-11-26 | Fortinet, Inc., A Delaware Corporation | Monitoring and dynamic tuning of target system performance |
US8140552B2 (en) * | 2008-09-19 | 2012-03-20 | International Business Machines Corporation | Method and apparatus for optimizing lead time for service provisioning |
US8041794B2 (en) * | 2008-09-29 | 2011-10-18 | Intel Corporation | Platform discovery, asset inventory, configuration, and provisioning in a pre-boot environment using web services |
US9875141B2 (en) * | 2008-10-01 | 2018-01-23 | Microsoft Technology Licensing, Llc | Managing pools of dynamic resources |
US8271974B2 (en) | 2008-10-08 | 2012-09-18 | Kaavo Inc. | Cloud computing lifecycle management for N-tier applications |
US7587718B1 (en) * | 2008-10-31 | 2009-09-08 | Synopsys, Inc. | Method and apparatus for enforcing a resource-usage policy in a compute farm |
US8819106B1 (en) | 2008-12-12 | 2014-08-26 | Amazon Technologies, Inc. | Managing distributed execution of programs |
US9880877B2 (en) | 2009-01-22 | 2018-01-30 | International Business Machines Corporation | Methods for rule-based dynamic resource adjustment for upstream and downstream processing units in response to an intermediate processing unit event |
US8296419B1 (en) * | 2009-03-31 | 2012-10-23 | Amazon Technologies, Inc. | Dynamically modifying a cluster of computing nodes used for distributed execution of a program |
US8839254B2 (en) * | 2009-06-26 | 2014-09-16 | Microsoft Corporation | Precomputation for data center load balancing |
US10877695B2 (en) | 2009-10-30 | 2020-12-29 | Iii Holdings 2, Llc | Memcached server functionality in a cluster of data processing nodes |
US11720290B2 (en) | 2009-10-30 | 2023-08-08 | Iii Holdings 2, Llc | Memcached server functionality in a cluster of data processing nodes |
US8656019B2 (en) * | 2009-12-17 | 2014-02-18 | International Business Machines Corporation | Data processing workload administration in a cloud computing environment |
DE102010029209B4 (en) | 2010-05-21 | 2014-06-18 | Offis E.V. | A method for dynamically distributing one or more services in a network of a plurality of computers |
JP5417287B2 (en) | 2010-09-06 | 2014-02-12 | 株式会社日立製作所 | Computer system and computer system control method |
JP4982600B2 (en) * | 2010-09-24 | 2012-07-25 | 株式会社日立製作所 | Resource lending method and resource lending system |
US8849469B2 (en) | 2010-10-28 | 2014-09-30 | Microsoft Corporation | Data center system that accommodates episodic computation |
US9253016B2 (en) | 2010-11-02 | 2016-02-02 | International Business Machines Corporation | Management of a data network of a computing environment |
US9081613B2 (en) * | 2010-11-02 | 2015-07-14 | International Business Machines Corporation | Unified resource manager providing a single point of control |
US9063738B2 (en) | 2010-11-22 | 2015-06-23 | Microsoft Technology Licensing, Llc | Dynamically placing computing jobs |
US8566838B2 (en) | 2011-03-11 | 2013-10-22 | Novell, Inc. | Techniques for workload coordination |
US9047126B2 (en) | 2011-04-06 | 2015-06-02 | International Business Machines Corporation | Continuous availability between sites at unlimited distances |
US9450838B2 (en) | 2011-06-27 | 2016-09-20 | Microsoft Technology Licensing, Llc | Resource management for cloud computing platforms |
US9595054B2 (en) | 2011-06-27 | 2017-03-14 | Microsoft Technology Licensing, Llc | Resource management for cloud computing platforms |
US9058304B2 (en) | 2011-06-30 | 2015-06-16 | International Business Machines Corporation | Continuous workload availability between sites at unlimited distances |
US9251033B2 (en) * | 2011-07-07 | 2016-02-02 | Vce Company, Llc | Automatic monitoring and just-in-time resource provisioning system |
US9158590B2 (en) * | 2011-08-08 | 2015-10-13 | International Business Machines Corporation | Dynamically acquiring computing resources in a networked computing environment |
US8898291B2 (en) | 2011-08-08 | 2014-11-25 | International Business Machines Corporation | Dynamically expanding computing resources in a networked computing environment |
US9378045B2 (en) * | 2013-02-28 | 2016-06-28 | Oracle International Corporation | System and method for supporting cooperative concurrency in a middleware machine environment |
US9712599B2 (en) | 2011-10-03 | 2017-07-18 | International Business Machines Corporation | Application peak load processing |
US8949429B1 (en) * | 2011-12-23 | 2015-02-03 | Amazon Technologies, Inc. | Client-managed hierarchical resource allocation |
US9280394B2 (en) | 2012-02-03 | 2016-03-08 | International Business Machines Corporation | Automatic cloud provisioning based on related internet news and social network trends |
KR20130101693A (en) * | 2012-03-06 | 2013-09-16 | 삼성전자주식회사 | Method and apparatus for power management in virtualization system using different operation system |
US9009241B2 (en) | 2012-03-30 | 2015-04-14 | International Business Machines Corporation | Determining crowd topics from communications in a focus area |
US9641449B2 (en) | 2012-05-22 | 2017-05-02 | International Business Machines Corporation | Variable configurations for workload distribution across multiple sites |
EP2883140A1 (en) * | 2012-08-07 | 2015-06-17 | Advanced Micro Devices, Inc. | System and method for tuning a cloud computing system |
US9952879B2 (en) | 2012-08-30 | 2018-04-24 | Microsoft Technology Licensing, Llc | Application pre-layout in byte-addressable persistent random access memory |
US9740500B2 (en) | 2012-08-30 | 2017-08-22 | Microsoft Technology Licensing, Llc | Layout system for operating systems using BPRAM |
US9384051B1 (en) | 2012-09-14 | 2016-07-05 | Emc Corporation | Adaptive policy generating method and system for performance optimization |
US9052952B1 (en) * | 2012-09-14 | 2015-06-09 | Emc Corporation | Adaptive backup model for optimizing backup performance |
SE537197C2 (en) * | 2012-10-05 | 2015-03-03 | Elastisys Ab | Method, node and computer program to enable automatic adaptation of resource units |
US9092205B2 (en) * | 2012-10-25 | 2015-07-28 | International Business Machines Corporation | Non-interrupting performance tuning using runtime reset |
US9378064B2 (en) * | 2012-11-15 | 2016-06-28 | Bank Of America Corporation | Orchestration management of information technology |
US9038068B2 (en) | 2012-11-15 | 2015-05-19 | Bank Of America Corporation | Capacity reclamation and resource adjustment |
US9038086B2 (en) * | 2012-11-15 | 2015-05-19 | Bank Of America Corporation | End to end modular information technology system |
US9092750B2 (en) * | 2013-02-22 | 2015-07-28 | International Business Machines Corporation | Rapidly optimizing staffing levels in a ticketing system using simulation |
US9087310B2 (en) * | 2013-02-22 | 2015-07-21 | International Business Machines Corporation | Optimizing staffing levels with reduced simulation |
GB2512847A (en) | 2013-04-09 | 2014-10-15 | Ibm | IT infrastructure prediction based on epidemiologic algorithm |
GB2517195A (en) * | 2013-08-15 | 2015-02-18 | Ibm | Computer system productivity monitoring |
EP3039560A1 (en) * | 2013-08-30 | 2016-07-06 | Hewlett Packard Enterprise Development LP | Maintain a service on a cloud network based on a scale rule |
US9851726B2 (en) | 2013-09-04 | 2017-12-26 | Panduit Corp. | Thermal capacity management |
GB2520972A (en) | 2013-12-05 | 2015-06-10 | Ibm | Workload management |
US9686207B2 (en) * | 2014-01-29 | 2017-06-20 | Vmware, Inc. | Application service level objective aware demand estimation |
US10484470B2 (en) | 2014-05-09 | 2019-11-19 | International Business Machines Corporation | Peak cyclical workload-based storage management in a multi-tier storage environment |
US9933804B2 (en) | 2014-07-11 | 2018-04-03 | Microsoft Technology Licensing, Llc | Server installation as a grid condition sensor |
US10234835B2 (en) | 2014-07-11 | 2019-03-19 | Microsoft Technology Licensing, Llc | Management of computing devices using modulated electricity |
US10581756B2 (en) * | 2014-09-09 | 2020-03-03 | Microsoft Technology Licensing, Llc | Nonintrusive dynamically-scalable network load generation |
US11138537B2 (en) | 2014-09-17 | 2021-10-05 | International Business Machines Corporation | Data volume-based server hardware sizing using edge case analysis |
US10318896B1 (en) * | 2014-09-19 | 2019-06-11 | Amazon Technologies, Inc. | Computing resource forecasting and optimization |
US9998393B2 (en) | 2015-03-04 | 2018-06-12 | International Business Machines Corporation | Method and system for managing resource capability in a service-centric system |
US9984141B2 (en) | 2015-08-21 | 2018-05-29 | International Business Machines Corporation | Inferring application type based on input-output characteristics of application storage resources |
US10171572B2 (en) | 2016-01-20 | 2019-01-01 | International Business Machines Corporation | Server pool management |
JP6640025B2 (en) * | 2016-05-27 | 2020-02-05 | 本田技研工業株式会社 | Distributed processing control system and distributed processing control method |
US10091904B2 (en) * | 2016-07-22 | 2018-10-02 | Intel Corporation | Storage sled for data center |
JP6787032B2 (en) | 2016-10-18 | 2020-11-18 | 富士通株式会社 | Control devices, control methods, and control programs |
KR102610537B1 (en) * | 2016-11-10 | 2023-12-06 | 삼성전자주식회사 | Solid state drive device and storage system having the same |
US10620930B2 (en) | 2017-05-05 | 2020-04-14 | Servicenow, Inc. | Software asset management |
US11210133B1 (en) * | 2017-06-12 | 2021-12-28 | Pure Storage, Inc. | Workload mobility between disparate execution environments |
KR102062037B1 (en) | 2018-05-16 | 2020-01-03 | 국민대학교산학협력단 | Apparatus and method of providing a cloud-based batch service |
US10754987B2 (en) * | 2018-09-24 | 2020-08-25 | International Business Machines Corporation | Secure micro-service data and service provisioning for IoT platforms |
US20220038390A1 (en) * | 2018-09-25 | 2022-02-03 | Nec Corporation | Design management apparatus, control method thereof, program and management system |
US20200184366A1 (en) * | 2018-12-06 | 2020-06-11 | Fujitsu Limited | Scheduling task graph operations |
TWI706246B (en) * | 2018-12-27 | 2020-10-01 | 技嘉科技股份有限公司 | Power mode management system, method for providing power mode parameter combination, method for updating power mode parameter combination, computer software and storage medium |
US11301138B2 (en) * | 2019-07-19 | 2022-04-12 | EMC IP Holding Company LLC | Dynamic balancing of input/output (IO) operations for a storage system |
US11055196B1 (en) | 2020-06-12 | 2021-07-06 | Bank Of America Corporation | System and method for optimizing technology stack architecture |
US10958523B1 (en) | 2020-07-28 | 2021-03-23 | Bank Of America Corporation | Consistent deployment of monitoring configurations on multiple computing systems |
US11188437B1 (en) | 2020-07-30 | 2021-11-30 | Bank Of America Corporation | Remote deployment of monitoring agents on computing systems |
CN112156453B (en) * | 2020-10-21 | 2022-06-03 | 腾讯科技(深圳)有限公司 | Example adaptive adjustment method, apparatus, computer readable storage medium and device |
US11368539B1 (en) * | 2021-05-27 | 2022-06-21 | International Business Machines Corporation | Application deployment in a multi-cluster environment |
KR20230053424A (en) | 2021-10-14 | 2023-04-21 | 주식회사 꾸미다 | Distributed and Collaborative Container Platform |
KR20230088109A (en) | 2021-12-10 | 2023-06-19 | 국방과학연구소 | Apparatus, method, computer-readable storage medium and computer program for detecting malfunction of virtual network functions based on unsupervised learning |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5878224A (en) * | 1996-05-24 | 1999-03-02 | Bell Communications Research, Inc. | System for preventing server overload by adaptively modifying gap interval that is used by source to limit number of transactions transmitted by source to server |
US6269078B1 (en) * | 1997-04-04 | 2001-07-31 | T. V. Lakshman | Method and apparatus for supporting compressed video with explicit rate congestion control |
US6006196A (en) * | 1997-05-01 | 1999-12-21 | International Business Machines Corporation | Method of estimating future replenishment requirements and inventory levels in physical distribution networks |
US6516350B1 (en) * | 1999-06-17 | 2003-02-04 | International Business Machines Corporation | Self-regulated resource management of distributed computer resources |
US6256773B1 (en) * | 1999-08-31 | 2001-07-03 | Accenture Llp | System, method and article of manufacture for configuration management in a development architecture framework |
US20020152305A1 (en) * | 2000-03-03 | 2002-10-17 | Jackson Gregory J. | Systems and methods for resource utilization analysis in information management environments |
US6957433B2 (en) | 2001-01-08 | 2005-10-18 | Hewlett-Packard Development Company, L.P. | System and method for adaptive performance optimization of data processing systems |
US7350186B2 (en) * | 2003-03-10 | 2008-03-25 | International Business Machines Corporation | Methods and apparatus for managing computing deployment in presence of variable workload |
US7039559B2 (en) * | 2003-03-10 | 2006-05-02 | International Business Machines Corporation | Methods and apparatus for performing adaptive and robust prediction |
-
2003
- 2003-03-10 US US10/384,973 patent/US7350186B2/en active Active
- 2003-08-29 CN CNB038261243A patent/CN100377094C/en not_active Expired - Lifetime
- 2003-08-29 KR KR1020057014631A patent/KR100826833B1/en active IP Right Grant
- 2003-08-29 CA CA2515470A patent/CA2515470C/en not_active Expired - Lifetime
- 2003-08-29 EP EP03749287A patent/EP1602031A2/en not_active Ceased
- 2003-08-29 WO PCT/US2003/027304 patent/WO2004081789A2/en active Application Filing
- 2003-08-29 MX MXPA05009648A patent/MXPA05009648A/en active IP Right Grant
- 2003-08-29 AU AU2003268329A patent/AU2003268329A1/en not_active Abandoned
- 2003-08-29 JP JP2004569411A patent/JP2006520027A/en active Pending
-
2004
- 2004-03-08 TW TW093106051A patent/TWI308269B/en active
-
2007
- 2007-06-15 US US11/763,726 patent/US8386995B2/en active Active
Also Published As
Publication number | Publication date |
---|---|
US20040181794A1 (en) | 2004-09-16 |
TWI308269B (en) | 2009-04-01 |
CN100377094C (en) | 2008-03-26 |
WO2004081789A3 (en) | 2005-06-30 |
EP1602031A2 (en) | 2005-12-07 |
JP2006520027A (en) | 2006-08-31 |
WO2004081789A2 (en) | 2004-09-23 |
CA2515470A1 (en) | 2004-09-23 |
AU2003268329A1 (en) | 2004-09-30 |
MXPA05009648A (en) | 2005-10-20 |
CN1751294A (en) | 2006-03-22 |
US20070240162A1 (en) | 2007-10-11 |
US7350186B2 (en) | 2008-03-25 |
KR20060023951A (en) | 2006-03-15 |
TW200506598A (en) | 2005-02-16 |
KR100826833B1 (en) | 2008-05-06 |
AU2003268329A8 (en) | 2004-09-30 |
US8386995B2 (en) | 2013-02-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2515470C (en) | Methods and apparatus for managing computing deployment in presence of variable workload | |
US7039559B2 (en) | Methods and apparatus for performing adaptive and robust prediction | |
US11656915B2 (en) | Virtual systems management | |
JP7355404B2 (en) | Automatic tuner for embedding cloud microservices | |
US9921809B2 (en) | Scaling a cloud infrastructure | |
US20150378786A1 (en) | Physical resource allocation | |
KR20040062941A (en) | Automatic data interpretation and implementation using performance capacity management framework over many servers | |
US20180351816A1 (en) | Methods and apparatus for parameter tuning using a cloud service | |
Breitgand et al. | An adaptive utilization accelerator for virtualized environments | |
US10778785B2 (en) | Cognitive method for detecting service availability in a cloud environment | |
Giannakopoulos et al. | Smilax: statistical machine learning autoscaler agent for Apache Flink | |
Bhaskar et al. | Dynamic allocation method for efficient load balancing in virtual machines for cloud computing environment | |
US20210349705A1 (en) | Performance sensitive storage system upgrade | |
Fang et al. | OCSO: Off-the-cloud service optimization for green efficient service resource utilization | |
US11307902B1 (en) | Preventing deployment failures of information technology workloads | |
US20220318058A1 (en) | Service load independent resource usage detection and scaling for container-based system | |
US20230051637A1 (en) | Adjusting data backups based on system details | |
Srivatsa et al. | A policy evaluation tool for multisite resource management | |
Zhang et al. | Tuning adaptive computations for the performance improvement of applications in JEE server | |
Procaccianti et al. | VU Research Portal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request | ||
MKEX | Expiry |
Effective date: 20230829 |