WO2005034464A1

WO2005034464A1 - Overload-protected service portal, method and computer program comprising program code means and computer program product for protecting an administration program for service programs from overload

Info

Publication number: WO2005034464A1
Application number: PCT/EP2004/051902
Authority: WO
Inventors: Stefan Augustin; Thorbjörn Hansen; Nicole Wengatz
Original assignee: Siemens Aktiengesellschaft
Priority date: 2003-09-29
Filing date: 2004-08-25
Publication date: 2005-04-14
Also published as: DE10345258A1

Abstract

The invention relates to an overload protection feature for an administration program or service portal, into which service programs are integrated. The service portal comprises an overload-protected service program that is adapted to determine load data and that is integrated into the service portal as one of the service programs. The service portal can be a web portal and the service programs can be web services. Load data preferably describe a memory usage, a CPU load, a number of user requests and/or a number of logged in users.

Description

description

Service portal with overload protection, method and computer program with program code means and computer program product for overload protection for an administration program for service programs

The invention relates to overload protection for an administration program for service programs.

Overload protection with corresponding overload protection mechanisms for a single (service) program is known from the prior art.

Such overload protection in the prior art has a task to ensure that the individual (service) program runs stably.

As a rule, the overload protection determines current load data of the individual (service) program, such as a memory usage or a CPU load, which current load data describe a current load / load state of the (service) program.

If determined load data exceed a predefinable limit, a load state of the individual (service) program "overload" is reached and an overload protection mechanism is triggered.

Such an overload protection mechanism can be, for example, a warning message or an action to counteract the overload, such as an access refusal.

If there is no such overload protection in a (service) program, the situation is usually high when the load is high. the (service) program takes very long response times.

There is also a risk of program crashes or technical error messages, which require further work with the

Make (service) program impossible (e.g. "Internal Server Error" for solutions based on Java Servlets).

An administrative program or a service portal for service programs, such as a so-called web portal, in which numerous application / service programs (applications) are integrated and available to a user, is known from [2].

Web portals

Web portals, portals for short, are known from [1] and [2].

Such a portal known from [1] or [2] serves the

Purpose to open a common, personalized access to data, expertise and applications, in the following subsumed under the common term "application" (application).

Shortened and simplified, but expressed clearly, portals are thus comparable to "digital doors" as access to a "virtual, abstract space", namely to the applications, i.e. to the data, expertise and applications.

Basic functions of a web portal

To achieve this purpose, a portal must fulfill various functions, so-called basic functions: 1. Control and navigation

The first function is a relief or control when searching for content. The term navigation is used for this.

This does not only include a surface search, i. H. not only the search for the specified keyword in the file name, but ideally also content should be included in the search.

In most cases it also makes sense to allow a suitable combination of search criteria in order to enable the user to specify the search more precisely.

External (web) and internal files should be included in the search. Accordingly, a web browser is suitable as the user interface.

2. Data integration and content management

Since the navigation should provide information from different sources, this requires the integration of different contents. This is summarized under the term content management.

Content management is not only the provision of content from various sources, but this function should also ensure that the content is used efficiently.

Sophisticated interface technology is required here to pursue this goal with appropriate means. Interface technologies such as B. Unified Content API (Application Programming Interface) support all common tools for the development of web environments and thus help to keep this effort limited. 3. Personalization

Information overflow is one of the biggest problems on the Internet. Search engines and search catalogs offer thousands of information. Therefore, the efficient use of information is only possible if a pre-selection has been made. This filter function should take over the personalization.

Accordingly, personalization represents an individual design of the website. The user should only be confronted with the personally relevant information and the applications he requires.

This can happen based on the role or profile of each user. A customer's profile can be created in two different ways: from data that he explicitly specifies or that already exists to the administrator of the respective portal and - from data that is implicit, e.g. B. from surfing behavior. The method for collecting such data is called data mining.

In order to be able to offer the user preferred products, information and applications on the basis of this recorded data, so-called rule-based methods are used, i.e. H. these procedures are based on predefined rules, such as the assignment of complementary products (supplementary products). These methods can be refined based on filtering methods. Metadata, i.e. age, place of residence, status etc. of the user are used to derive possible personal preferences.

4. Notification and notification The portal should also be able to transmit signals relevant to the user. This function is called notification.

This is understood to be an automated technique that the user when the defined parameters are undershot or exceeded, i. H. relevant aspects, notified. It can either be used to meet certain deadlines or milestones (one-dimensional in terms of time), or it can also generally monitor business parameters to maintain business processes (general, multi-dimensional).

5. Knowledge management or knowledge management

However, these functions described so far only make sense in an environment of suitable information. The fifth function of the portals, knowledge management, is necessary to ensure that they are available at the right place and at the right time.

Inside and outside of a company, relevant knowledge is created daily for the portal user. The appropriate structuring and individual supply of this knowledge creates a cost and innovation-relevant advantage.

Methods from the area of data and information management are used: On the one hand, a function for managing profiles of users and interest groups is required. These should be structured according to knowledge, skills and current fields of activity. On the other hand, functions for localizing and structuring existing knowledge are required. A so-called content catalog can serve this purpose, which represents a kind of site plan and is intended to embody a network of relationships between the information. Such "knowledge networks" make the search for what is actually suitable easier and more effective.

6. Workflow and workflow management

Another efficiency advantage is also created by a suitable workflow design - the so-called workflow management.

With an appropriate system, asynchronous, i. H. coordinated group work that is not coordinated. A simple task here is, for example, the internal formalized communication of a company. It is much more difficult to automatically assign work and tasks within a group, which should be based on the status, workload and skills of the individual employee.

The difficulty in modeling and maintaining complex workflows becomes clear here. For this reason, this function is often limited to work processes with a high routine content, and existing solutions are usually integrated into the portal instead of implementing them anew, i.e. to implement in software.

7. Application integration

This is the seventh function of an ideal portal.

A distinction is usually made between integration on the customer side and one on the server side.

The former is limited to displaying different, mutually independent applications. Since these do not influence each other, this integration is easy to implement. The integration on the server side makes the former unnecessary. It can only be done on the basis of a web application server. An attempt is made to implement middleware that is seen as a uniform interface and is separate from the actual system platform, the application server. This abstracting means that the portal can also be transferred to more powerful systems at any time.

This can be done by a server that processes so-called servlets, which allow the construction of dynamic HTML pages, or can change the websites according to a Java program by using JSPs (Java Server Pages).

This possibility of application integration also enables the development of a new application logic, but this is not an essential part of portals.

8. Infrastructure services

The only thing missing is a basis on which all other functions are based. These include e.g. b. the web infrastructure services mentioned above.

Application servers and other services only make a single sign-on possible. H. that the user only has to identify once for all the services offered. This basic function allows all the functions mentioned above to become one unit, from which all others are available.

Overload protection for web portals

The enormous increase in the use of the Internet and the associated enormous increase in (user) access to the Internet, in particular to web portals, which provide services through integrated applications on the Internet, make overload protection increasingly important. Various designs of overload protection in web portals are known from the prior art.

As with a web portal with integrated applications, an overload contactor can be implemented or integrated individually in each application.

Such individual integration of overload protection in an application is described in the prior art,

However, such an overload protection consisting of individual protection mechanisms in a web portal has various disadvantages.

Because of the numerous individual implementations of overload protection in the applications, there are high development costs for the web portal.

Furthermore, load data from one application are not available to other applications or their overload protection mechanisms.

The invention is therefore based on the object of specifying an overload protection for a service portal with integrated applications, which overload protection can be easily integrated into the service portal, offers safe and comprehensive protection against overloading the service portal and the integrated applications and, moreover, can be implemented inexpensively ,

This task is solved by the service portal with overload protection, by the method as well as the computer program with program code means and the computer program product for overload protection for an administration program for service programs with the features according to the respective independent patent claim. The service portal with overload protection refers to the service portal and service programs that are integrated in the service portal, as well as an overload protection service program that is set up to determine load data.

The invention is further characterized in that the overload protection service program is integrated into the service portal as one of the service programs.

The method for overload protection for an administration program for service programs which are integrated into the administration program uses an overload protection service program which is set up for determining load data and which is integrated into the service portal as one of the service programs.

In the overload protection method according to the invention for the

Administration program determines the overload protection service program load data of at least some of the service programs and uses the determined load data to assess a load status of the administration program.

In the case of the invention, a service program is understood to mean a program which offers all types of information and service services, for example only the provision of information.

From an illustrative point of view, the invention deals with a portal - also functionally referred to as an administration program - in which an overload protection or overload protection mechanism is or is integrated as a service for the portal or administration program and for services integrated in the portal or administration program , In addition to the overload protection of the entire portal, the individual portal services (portal applications) can also be protected. They can use the portal's overload protection without having to implement their own overload protection.

The approach according to the invention is characterized by numerous advantages which it can achieve:

- Response times from the portal and portal services (portal applications) are short. The portal and portal services (portal applications) are stable. In particular, crashes caused by overloading and technical errors in the portals and portal services (portal applications) are avoided, which make further work impossible. If the portal is overloaded, the work of users who are already logged in is not affected. If a portal service is overloaded (portal application), the work of a user in the portal and in other portal services (portal applications) is not impaired. Overload metrics, as described below by way of example, can be configured in many ways.

Overload metrics

In the case of the overload protection or associated overload protection mechanisms according to the invention, the following metrics can be implemented, for example, as part of further training: limitation of memory use, limitation of a CPU load, limitation per time unit of user requests that can be processed, limitation (simultaneous or cumulative) of users logged into the portal ,

Embodiments and implementations of these metrics are briefly described below: Limitation of food use:

Every time a user wants to log into the portal, it is checked whether the current memory requirement is above a configurable threshold. If this is the case, access is denied and the user receives an error message asking him to try again later.

Limiting the CPU load: Every time a user wants to log into the portal, it is checked whether the current CPU utilization is above a configurable threshold. If this is the case, access is denied and the user receives an error message asking him to try again later.

Limitation of user requests to be processed per minute:

Every time a user wants to log into the portal, it is checked whether the number of user requests already received per minute is above a configurable threshold. If this is the case, access is denied and the user receives an error message asking him to try again later.

A variant of this metric is that not all requests are counted, only the login requests.

Limitation of users logged into the portal at the same time

Every time a user wants to log into the portal, it is checked whether the number of users currently logged in is above a configurable threshold. If this is the case, access is denied and the user receives an error message asking him to try again later. It can apply to all metrics that the configuration parameters can be changed at runtime of the portal. Users who are already logged into the portal can continue their work even under load. New login attempts are rejected in the event of an overload. The system administrator is notified in the event of an overload, for example by email. The current values of all metrics are displayed so that he can identify the bottlenecks of the system and, if necessary, adapt the system (eg add more memory).

Further advantages of the approach according to the invention are mentioned below:

- Bottlenecks are easy to identify and fix, for example by a system administrator.

- Parameters for the individual portal applications or portal components can be specified separately.

- Overload criteria and data are available to other applications. If, for example, a portal accesses a login service and the login service is overloaded, the overload protection can report this to the portal so that the portal does not use unnecessary resources and remains stable.

The computer programs according to the invention, each with program code means, are set up to carry out all steps according to the respective method according to the invention when the respective program is executed on a computer.

The computer program products, each with program code means stored on a machine-readable carrier, are set up to carry out all steps according to the respective method according to the invention when the respective program is executed on a computer. The computer programs, each with program code means, set up to carry out all steps in accordance with the respective inventive method when the respective program is executed on a computer, and the computer program products, each with program code means stored on a machine-readable medium, set up for all Performing steps according to the respective inventive method when the respective program is executed on a computer are particularly suitable for carrying out the method according to the invention or its further developments explained below.

Preferred developments of the invention result from the dependent claims.

The further developments described below relate both to the methods and to the computer programs and computer program products.

The invention and the further developments described below can be implemented both in software and in hardware, for example using a special electrical circuit.

Furthermore, an implementation of the invention or a further development described below is possible by means of a computer-readable storage medium on which the computer program with program code means which carries out the invention or further development is stored.

The invention or any further development described below can also be implemented by a computer program product which has a storage medium on which the computer program with program code means which carries out the invention or further development is stored. As part of a further development of the invention in the context of a WWW, the service portal can be a web portal and one of the service programs can be a web service.

Furthermore, the load data can describe a load on the service portal or a load on at least one of the service programs.

In addition, the load data can describe a memory usage, a CPU load, a number of user requests and / or a number of logged in users. These can be taken into account or configured in the context of corresponding overload metrics (see description of the overload metrics above).

To limit the load, it is then expedient to specify a limit for the load data.

It is also expedient that the overload protection service program triggers an overload protection mechanism, in particular a warning and / or an action counteracting an overload, if determined load data exceed the predetermined limit and the overload is thereby determined.

If, for example, a service program (application) is overloaded, an error message appears as a warning as soon as another user wants to use the application. In addition, the affected user can be informed of the overload, for example with the instruction to log into the portal at a later time.

The overload protection service program can be executed if you log in to the service portal or access one of the service programs. Embodiments of the invention are shown in figures and are explained below.

Show it

Figure 1 sketch showing an overload protection control at a login time according to a first embodiment;

Figure 2 sketch showing an overload protection control for an application according to a second embodiment;

FIG. 3 sketch with an error page which is displayed to a user in the event of an established overload in accordance with a first and a second exemplary embodiment.

First embodiment: overload protection control of a web portal at the time of login to the web portal (Fig.l)

The subject of the first exemplary embodiment is an overload control of a web portal at the time of login.

Fig.l schematically shows this overload control 100.

The overload control 100 according to the first exemplary embodiment or according to FIG. 1 is characterized in that a service 120 of the web portal 110, in this case a portal login service 120, accesses an overload protection service 130.

When a login attempt 140 by a user 142 into the web portal, a login request 141, an HTTP request, is sent from a browser of a user 142 to a portal server (1). In the web portal 110 according to the first exemplary embodiment, the request 141 is processed by the portal login service 120 integrated in the web portal 110. For this purpose, the request 141 is forwarded to the portal login service 120 (2).

As part of this processing of request 141, an overload check of web portal 110 takes place.

For this purpose, the portal login service 120 makes use of a service of the overload protection service (overload module) 130 integrated in the web portal 110 (FIG. 3).

The overload module 130 integrated in the web portal 110 regularly determines 150 load data from applications 151 and computers 152 integrated in the web portal 110, in this case a CPU load.

With the current request 141, the overload module 130 now checks whether the current CPU utilization of the web portal 110 is above a configurable threshold value and thus the web portal 110 is currently overloaded.

The check result 160 is sent to the portal login service

120 returned (4).

The portal login service 120 acts 170 according to the check result 160:

If yes (5a), i.e. in the event of an overload, the user gets

142 a corresponding error page 173 is displayed 171 (FIGS. 3, 173).

In the other case (5b), i.e. with no overload, the

User 142 successfully logs into portal 110 172,

174th

Users who are already logged in are not affected by an overload that may have been detected in a current request. Second embodiment: overload protection control for an application in a web portal (Fig. 2)

The second exemplary embodiment relates to an overload control for applications of a web portal.

2 schematically shows this overload control 200.

The overload control 200 according to the second exemplary embodiment or according to FIG. 2 is characterized in that a web portal 210 accesses an overload protection service (overload module) 220 (cf. first exemplary embodiment: access by the portal login service) the overload module).

In the second embodiment of the web portal 210, in this case in a Siemens Employee Portal, a service or an application “Employee Services” (ES) 230 is integrated.

The 'employee service' 230 runs on an external application server.

If a portal user 242 activates the application "Employee Service" 230 or requests access to the "Employee Service" 240 from the web portal (1), the overload of the ES 230 is checked.

For this purpose, the web portal 210 makes use of a service of the overload protection service (overload module) 220 integrated into the web portal in a modular manner in claim 241 (2).

The overload module 220 integrated in the web portal 210 regularly determines load data 250 from applications 251 and computers 252 integrated in the web portal 210, in this case a CPU load, and also from the ES 230. If the overload module receives the current request 241 from the web portal 210 for an overload check of the ES 230, this determines the current load data 250 of the ES 230 and checks whether the current CPU utilization of the ES 230 is above a configurable threshold value.

The check result 260 is returned 261 to the web portal 210 (FIG. 3).

The web portal 210 acts 270 according to the check result 260, 271 and 272:

If yes (4a), i.e. in the event of an overload, the user 242 is shown 272 a corresponding error page 273 (FIGS. 3, 273).

In the other case (4b) 272, ie with no overload, the ES 230 is started 274.

The following publications are cited in this document:

[1] Overload protection for a single program, available on May 12, 2003 at: ht p: // ww .corrputerwoche.de / hef archiv / 2002/20021011 / a80 10_97l4.html

[2] Definition of the Internet portal, available on April 16, 2003 at: http: //www.mathematik. uni- ulm. de / sai / ws01 / portalsem / bvb / Arbeit / Internetportale% 20-% 20History,% 20Kriterionen% 20und% 20Funktionen.pdf.

Claims

claims

1. service portal with overload protection,

- with the service portal and service programs that are integrated into the service portal,

- With an overload protection service program, which is set up to determine load data, characterized in that

- The overload protection service program is integrated into the service portal as one of the service programs.

2. Service portal with overload protection according to claim 1,

- where the service portal is a web portal.

3. Service portal with overload protection according to one of the preceding claims,

- where one of the service programs is a web service.

4. Service portal with overload protection according to one of the preceding claims, in which the load data describes a load on the service portal or a load of at least one of the service programs.

5. Service portal with overload protection according to one of the preceding claims, in which the load data describe a memory usage, a CPU load, a number of user requests and / or a number of logged in users.

6. Service portal with overload protection according to one of the preceding claims, - in which a limit is specified for the load data.

7. Service portal with overload protection according to the preceding claim, in which the overload protection service program triggers an overload protection mechanism, in particular a warning and / or an action counteracting an overload, if determined load data exceed the predetermined limit and the overload is thereby determined.

8. Service portal with overload protection according to one of the preceding claims, in which the overload protection service program is executed when a login to the service portal or access to one of the service programs takes place.

9.Procedure for overload protection for an administration program for service programs which are integrated into the administration program, and with an overload protection service program which is set up for determining load data and which is integrated into the service portal as one of the service programs in which the overload protection Service program load data determined by at least some of the service programs - and in which the overload protection service program uses the determined load data to assess a load status of the administration program.

10. A computer program product comprising a computer-readable storage medium on which a program is stored which enables a computer, after it has been loaded into a memory of the computer, to carry out all the steps according to claim 9.

11. Computer-readable storage medium on which a program is stored which enables a computer after it has been loaded into a memory of the computer to carry out all steps according to claim 9.

12. Computer program with program code means to carry out all steps according to claim 9 when the program is executed on a computer.

13. Computer program with program code means according to claim 12, which are stored on a computer-readable data carrier.