CN100435111C - Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment - Google Patents
Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment Download PDFInfo
- Publication number
- CN100435111C CN100435111C CNB2005100304987A CN200510030498A CN100435111C CN 100435111 C CN100435111 C CN 100435111C CN B2005100304987 A CNB2005100304987 A CN B2005100304987A CN 200510030498 A CN200510030498 A CN 200510030498A CN 100435111 C CN100435111 C CN 100435111C
- Authority
- CN
- China
- Prior art keywords
- program
- debugging
- pitching pile
- parallel
- function
- 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 - Fee Related
Links
Images
Abstract
The invention supplies multi-language multiplatform parallel debugging and performance analysis method under isomerization environment. The parallel debugging and performance analysis system is divided into monitoring end, concentrating watching process, and long distance running node. It realizes visualization performance analysis and debugging which includes the steps that front end code analysis, auto pile pitching, debugging support and performance analysis, and visualization user management. It supports isomerization parallel environment, multi-language and parallel platform of which architecture irrespective with the language and parallel platform has pile pitching module independent of the concrete programming language and parallel library. It supplies a new method and means for multi-language multiplatform parallel debugging and performance analysis under isomerization environment.
Description
Technical field
The present invention relates to a kind of parallel debugging and method for analyzing performance, belong to the debugging monitoring technique field in the high performance parallel computation.
Background technology
At present, the research of network high-performance computing environment is subjected to extensive attention, and the thinking and the way of realization of structure also have nothing in common with each other.According to the structure form, they can be divided into two relevant big classes of technology:
(1) cluster calculates.Adopt express network to connect one group of workstation or microcomputer, form a cluster, or on general network, seek dynamic Virtual Cluster of one group of idle processor formation, make it under intermediate management control, to provide to have the very high performance computing service of high performance-price ratio.
(2) grid computing.Its target is by computational resource, data source and other the equipment organization of unity management of internet with wide scope, forms the high-performance computing environment that can utilize mutually, cooperate with each other, and the user can use it as super giant machine of login.
No matter cluster calculates or grid computing, and its basic thought all is that the computational resource of various dispersions is integrated, thereby realizes parallel computation efficiently.Especially grid is as the trend of high-performance calculation development, and the system entails in its environment is the system of various isomeries, or is positioned under the different network architecture, or adopts different operating platforms.Complicacy in the high-performance computing environment that causes thus, the diversity of parallel programming language and parallel platform for the exploitation of high-quality concurrent software has caused bigger difficulty, is also had higher requirement for the debugging of concurrent program and performance monitoring analytical work.
Find that by literature search the debugging of traditional concurrent program and performance analysis tool calculate the ParaVision of institute etc. as dawn 2000 parallel debugging device DCDB and the Chinese Academy of Sciences, for the extensibility aspect of different platform weakness comparatively.And for the isomerism characteristics in cluster or the grid computing environment, the at present domestic related tool that does not also have maturation.
A few is provided by reasonable third party manufacturer except TotalView of Etnus etc. in the external similar tools, other general all is the operating system of the parallel machine exploitation of the specific model of producing at them that provided by parallel computer manufacturers such as SUN, IBM, HP, therefore these debugging acids compatible relatively poor for platform, and the instrument that has only provided the debugging of concurrent program, can not carry out performance evaluation well.Therefore the hope exploitation compatibility that the researchist of parallel computation both at home and abroad is urgent at present, comprehensive good effective parallel debugging and performance evaluation aid.
Summary of the invention
The objective of the invention is to, defective or deficiency at the prior art existence, parallel debugging and the method for analyzing performance of supporting multi-language multi-platform under the isomerous environment is provided for the user, unify data representation between heterogeneous nodes by the definition common type system, finish the dynamic debugging function, and the definition by process event is carried out detailed graphical analysis to the execution performance of concurrent program.
For reaching above-mentioned purpose, the present invention adopts following technical scheme:
Support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment, consider the isomerism of cluster system, from in logic local monitor end and long-range operation node being separated completely, parallel environment debugging and performance analysis system are divided into local monitor end, cluster finger daemon and long-range operation node three parts.
Simultaneously,, it is divided into four big steps successively, i.e. front end code analysis step, pitching pile step, debugging support and performance evaluation step and visual user management step automatically for realizing visual performance evaluation and debugging.
Wherein, front end code analysis step is carried out pre-service and morphology grammatical analysis to source code, generation complete syntax tree, and the tabulation that all User-Defined Functions are exported in analysis according to syntax tree, call relation between the function, and other a series of supplementarys, make and when pitching pile, can carry out cross reference analysis between the same engineering multifile, and the derivation of syntax tree and the reverse function of generating code are provided after pitching pile finishes.
Automatically the pitching pile step is that the code after analyzing is made amendment and packed, and inserts debugging code with logging program ruuning situation and collect performance parameter.It is finished at cluster finger daemon place.For realizing the independence to language and platform, we separate the specific implementation in pitching pile storehouse and the semanteme in pitching pile storehouse.In concrete pitching pile, will revise user's code according to the semanteme of pitching pile masterplate, but the realization of concrete masterplate, then can be by user's modification and expansion.The realization mechanism in this semanteme and the pitching pile storehouse of realizing separating is called the pitching pile template base, uses such pitching pile technology to provide good dirigibility and extensibility as the user.Based on the definition and the design of the pitching pile script of Program Semantics structure, the configuration file that the code analysis result generates has write down the function and the breakpoint information of whole procedure, is used to control whole pitching pile process.Can reach the design in the pitching pile storehouse in parallel storehouse at multilingual, pitching pile template mechanism is separated the specific implementation in pitching pile storehouse and the semanteme in pitching pile storehouse.
The debugging support mainly is on long-range parallel node the code behind the pitching pile to be moved with the performance evaluation step, carries out the static and dynamic debugging breakpoints of concurrent program simultaneously, and the performance parameter of collection procedure operational process is returned the local monitor end to these information transmission.
Visual user management step mainly is that the real-time status to each parallel node in the program operation process shows, and can carry out tree-like demonstration to the function call situation in the program operation process, the static breakpoint in simultaneously can the display routine operational process and dynamically breakpoint information and memory variable value.Visual performance evaluation can give patterned demonstration and analysis to the performance data in the program operation process, helps to find performance bottleneck or mistake in the program, thereby helps the optimization of program.Based on the visual performance evaluation technology of measuring, the assurance of pitching pile mechanism can the collection procedure operational process in a large amount of performance datas, be used for the performance evaluation of program.
Technical solution of the present invention is supported the isomery parallel environment, and can support multilingual and parallel platform on architecture.Itself and language reach the architecture of parallel platform independence, have the pitching pile template that is independent of concrete programming language and parallel storehouse.For supporting the parallel debugging and the performance evaluation of multi-language multi-platform that a kind of new ways and means is provided under the isomerous environment.
Further specify the present invention below in conjunction with drawings and Examples.
Description of drawings
Fig. 1 parallel environment debugging of the present invention and performance analysis system architectural schematic;
Fig. 2 Tongji University parallel debugging and performance analysis system are specifically implemented synoptic diagram;
Fig. 3 the inventive method process flow diagram.
Embodiment
Support the parallel debugging and the method for analyzing performance of multi-language multi-platform under a kind of isomerous environment, concurrent program debugging and visual performance analysis system mainly comprise the two large divisions under its isomerous environment: local monitor end and long-range parallel running node.The local monitor end mainly carries out the morphology grammatical analysis by the source program to input, receives concurrent program Debugging message and performance analysis data from long-range operation node simultaneously, then these information and data is analyzed and visual demonstration.Long-range operation node receives the result of local monitor end code analysis, according to the pitching pile template base program is carried out pitching pile, code behind the pitching pile produces certain Debugging message and performance analysis of program data after operation on the remote node, these information and data in real time send it back the local monitor end.Concurrent program debugging and performance analysis system under the isomerous environment based on dawn 3000 high performance parallel computation machines, have been constructed in national high performance computer engineering and technique center Tongji University branch center.
Tongji University's parallel debugging and performance analysis system are concrete to be implemented as shown in Figure 2: wherein,
The local monitor end is PC (personal computer), and Redhat Linux 9.0 operating systems have been installed, parallel environment: PVM3.4.3, and development environment: Kylix 3.0 enterprise versions are supported STL (STL).Long-range operation end is dawn 3000 high performance parallel computation machines, and AIX 4.3.3 operating system has been installed, parallel environment: PVM3.4.3, translation and compiling environment: GCC3.2.2 version.Mutually by the Socket communication.
At first set up an engineering, and engineering is added file for this reason: source code, header file and secondary file at the local monitor end.Then this engineering is carried out code analysis, the result of code analysis generates the syntax tree of source code, generates a configuration file simultaneously, and the function of pitching pile and the static breakpoint information of user's setting etc. wanted in record.
The local monitor end sends to the result of code analysis and configuration file the finger daemon of long-range parallel node then, send pitching pile and requirement for compiler simultaneously, finger daemon mainly is that the entrance and exit to program carries out function pitching pile, the pitching pile of parallel built-in function and the pitching pile of static breakpoint that pitching pile, user are realized, result behind the pitching pile has carried out some packings and modification to function in the source program, insert debugging code and performance tracking data, write down program static debugging breakpoints information simultaneously.
Code behind the pitching pile compiles at long-range parallel node then, generates executable file.
Then the local monitor end sends the request of executive routine to long-range operation node, local monitor end operation executable file, simultaneously the debugging of concurrent program is carried out in the dynamic debugging breakpoints order of sending according to the static breakpoint information of having established in the program and local monitor end.In the program operation process, the performance data in the long-range operation node while collection procedure operational process.
Long-range operation end sends it back the local monitor end with performance data by certain format.
The performance statistics that last local monitor end basis is collected, the dynamically current practice condition of display routine.In the operational process of whole procedure, the local monitor end can send debug command to executory program at any time, the execution of interruption and recovery routine.
Its process flow diagram as shown in Figure 3, concrete steps are described below:
1) beginning is at local monitor end loading routine source code;
2) the local monitor end carries out code analysis to source program, comprising lexical analysis and grammatical analysis;
3) result of code analysis generates the syntax tree of source program, by from source code, extracting some information, generate a configuration file, write down the file name that refers to therein, provide the function name of realization, the title of the variable that defines in the program and breakpoint information by the user;
4) syntax tree behind the code analysis and the configuration file cluster finger daemon that sends long-range operation node to carries out pitching pile work;
5) the automatic pitching pile device on the long-range parallel node is controlled whole pitching pile process according to the profile information that receives;
6) whole procedure is carried out pitching pile;
7) to the user function in the program and and line function carry out pitching pile;
8) the middle static breakpoint to program carries out pitching pile;
9) begin template in the porch of program insertion program, this template is used for the global variable of initialization self, and obtains a series of environmental variance, comprises the start time of program, function number, process number etc.;
10) withdraw from template in the exit of program insertion program, the internal memory of application when discharging debugging when this template is used for EOP (end of program), concluding time of logging program simultaneously, function number etc.;
11) to the pitching pile of user function.User function is meant the function that specific implementation is arranged in user program, and provides the function wrapping code by the user, the built-in function that calls voluntarily.The content of pitching pile comprises three aspects: the statement of function, realize and call pitching pile;
12) to the pitching pile of parallel built-in function to and the pitching pile of line function, owing to have us to need the information of collecting in the operational process in the parallel built-in function, therefore except inserting basic templates, also need this class function is further packed, insert some performance evaluation functions, process number, communication time and traffic when being used for the recording messages transmission;
13) before static breakpoint, insert the debugging breakpoints template;
14) behind static breakpoint, insert the debugging breakpoints template;
15) the program pitching pile as a result the time write down the outlet of program and entry position, start and end time and function number, document number etc.;
16) result of function pitching pile has not only write down the beginning and the post-set time of function, and has write down the communication information between the process, is used for the performance data of collection procedure operational process;
17) position of the outcome record of static breakpoint pitching pile debugging provides the code of debugging breakpoints to be used to collect Debugging message simultaneously;
18) syntax tree behind the pitching pile is reduced into the code of concurrent program;
19) pitching pile post code is carried out static state and dynamic debugging simultaneously in long-range parallel node operation, and performance analysis of program;
20) information and the performance analysis data of long-range parallel node after program debug passed the local monitor end back;
21) the local monitor end carries out patterned demonstration and analysis to the data that receive;
22) withdraw from.
Wherein, in the work of pitching pile, program begins/withdraws from pitching pile (step 9, step 10, step 15) automatically.Specifically describe as follows: the inlet insertion program in program begins template.In the program engineering for the C language, the main function is the entrance, and all personal code works of pitching pile will be placed in before the first statement in the main function.In this template, the global variable that needs initialization self, and obtain a series of environmental variance, the user can also add the setup code of oneself in this template in addition, collect the relevant information of some program run, comprise start time, process number, job number and function number of program etc.
Outlet insertion program in program withdraws from template.For the program of C language, all personal code works of pitching pile will be placed in after the last item statement of main function and before each bar return (returning) statement.In this template, discharge the internal memory apply for for debugging, and do some cleaning works, the user can also add the setup code of oneself in this template, collect the relevant information of some program run, comprises post-set time of program and function number etc.
Function pitching pile (step 11, step 12, step 16).Mainly comprise pitching pile to user function, to and the pitching pile of line function.Specifically describe as follows:
Pitching pile to user function.User function is meant the function that specific implementation is arranged in user program, and provides the function wrapping code by the user, the built-in function that calls voluntarily.The content of pitching pile comprises three aspects: one, add parameter to write down its particular location in program in function declaration.Two, begin place's insertion function at function and enter template, before function reaches each bar return statement at last, insert function respectively and withdraw from template.These two templates provide resolves point code so that dynamically arrange breakpoint in debugging, can insert the User Defined code, debug events and performance evaluation incident simultaneously.Three, because function declaration changes, need to revise all to this function calls, to guarantee to produce correct Debugging message.
To and the pitching pile of line function.Need the information of collecting in the operational process owing to have in the parallel built-in function, therefore except inserting basic templates, also need this class function is further packed.That is, in the pitching pile code, need to write down the parameters that needs are analyzed,, and send to server then according to the communication primitive of these parameter initialization correspondences.Carry out this message when transmitting communication calling former parallel built-in function, collect process number, communication time and traffic.
Static breakpoint pitching pile (step 13, step 14, step 17).Static statement breakpoint was provided with in program run in the past by the user, need and insert breakpoint front and back template during pitching pile before being set up the statement of breakpoint afterwards respectively, and add certain trigger condition on request.But the breakpoint code is provided in this template,, and can activates dynamically or cancel so that in debugging, can stop at the breakpoint place.
Claims (6)
1, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under a kind of isomerous environment, it is characterized in that may further comprise the steps:
1) front end code analysis, in the line code analysis of going forward side by side of local monitor end loading routine source code, generate the syntax tree and the configuration file of source program and be sent to the cluster finger daemon of long-range operation node, described configuration file has write down the function and the breakpoint information of whole procedure, is used to control whole pitching pile process;
2) automatic pitching pile, the cluster finger daemon of long-range operation node carries out pitching pile to this syntax tree;
3) debugging and performance evaluation, syntax tree behind the pitching pile is reduced into the code of concurrent program, and, carries out static state and dynamic debugging simultaneously in this long-range operation node operation, and performance analysis of program, pass information after the program debug and performance analysis data back the local monitor end;
4) visual user management, the local monitor end carries out patterned demonstration and analysis to the data that receive.
2, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment according to claim 1, it is characterized in that: described configuration file has write down the file name that refers to, provide the function name of realization, the title of the variable that defines in the program and breakpoint information by the user.
3, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment according to claim 1, it is characterized in that: described automatic pitching pile comprises program, function, static breakpoint pitching pile.
4, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment according to claim 3, it is characterized in that described program pitching pile may further comprise the steps:
(1) begin template in the porch of program insertion program, this template is used for the global variable of initialization self, and obtains environmental variance, comprises the start time of program, function number, process number;
(2) withdraw from template in the exit of program insertion program, the internal memory of application when discharging debugging when this template is used for EOP (end of program), logging program concluding time, function number simultaneously.
5, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment according to claim 3, it is characterized in that: described function pitching pile comprise to the pitching pile of user function and to and the pitching pile of line function.
6, support the parallel debugging and the method for analyzing performance of multi-language multi-platform under the isomerous environment according to claim 3, it is characterized in that: described static breakpoint pitching pile comprises insertion debugging breakpoints template before static breakpoint and behind the static breakpoint respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100304987A CN100435111C (en) | 2005-10-13 | 2005-10-13 | Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100304987A CN100435111C (en) | 2005-10-13 | 2005-10-13 | Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1949185A CN1949185A (en) | 2007-04-18 |
| CN100435111C true CN100435111C (en) | 2008-11-19 |
Family
ID=38018709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100304987A Expired - Fee Related CN100435111C (en) | 2005-10-13 | 2005-10-13 | Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100435111C (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101727316B (en) * | 2008-10-22 | 2013-01-02 | 上海超级计算中心 | Parallel program integrated development system suitable for clustering system as well as application implementation method thereof |
| JP5476834B2 (en) * | 2009-07-24 | 2014-04-23 | 株式会社リコー | Information processing apparatus, workflow system, workflow management method, program, and recording medium |
| CN101667156B (en) * | 2009-09-22 | 2012-09-26 | 中兴通讯股份有限公司 | Visual service flow debugging method and system |
| CN102023920B (en) * | 2010-10-27 | 2012-09-05 | 西安交通大学 | Method for gathering messages in remote parallel program debugging system based on tree form |
| CN102110052A (en) * | 2011-03-03 | 2011-06-29 | 华中科技大学 | Parallel acceleration method for dynamic analysis of program behavior |
| CN102789413B (en) * | 2011-05-23 | 2016-02-17 | 同济大学 | A kind of debug system of concurrent program and method |
| CN102902620B (en) * | 2011-12-23 | 2015-06-03 | 同济大学 | GDB (genome database) based heterogeneous computing and debugging environment realizing system |
| CN103455364B (en) * | 2013-09-05 | 2016-08-17 | 北京航空航天大学 | A kind of multi-core environment concurrent program Cache performance online obtains system and method |
| CN104252402B (en) * | 2014-09-05 | 2018-04-27 | 深圳创维数字技术有限公司 | A kind of program debugging method and device |
| CN104360936B (en) * | 2014-10-17 | 2017-02-15 | 中国人民解放军国防科学技术大学 | Barrier operation property analysis method based on triggering operation |
| CN106250179B (en) * | 2016-07-26 | 2019-10-25 | 北京北森云计算股份有限公司 | The method that realization system dynamic function intercepts extension is compiled by multilingual cloud |
| CN106250178B (en) * | 2016-07-26 | 2019-10-25 | 北京北森云计算股份有限公司 | The method that realization system dynamic function intercepts extension online is compiled by multilingual cloud |
| CN106445784B (en) * | 2016-09-27 | 2020-08-28 | 北京搜狐新动力信息技术有限公司 | Information monitoring method and device |
| CN108021495B (en) * | 2016-11-03 | 2021-05-14 | 北京计算机技术及应用研究所 | Code debugging method based on playback |
| CN107247671A (en) * | 2017-07-03 | 2017-10-13 | 郑州云海信息技术有限公司 | A kind of data capture method and device |
| CN115562962A (en) * | 2022-09-28 | 2023-01-03 | 北京百度网讯科技有限公司 | Method and device for analyzing model reasoning performance and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010056341A1 (en) * | 2000-03-15 | 2001-12-27 | Pennello Thomas J. | Method and apparatus for debugging programs in a distributed environment |
| CN1499363A (en) * | 2002-10-24 | 2004-05-26 | 国际商业机器公司 | Method and device for setting break point on regulating integrated executable program of isomeric system structure |
| US6941492B1 (en) * | 2002-02-05 | 2005-09-06 | Emc Corporation | Debugging tool for efficient switching between targets in a multi-processor environment |
-
2005
- 2005-10-13 CN CNB2005100304987A patent/CN100435111C/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010056341A1 (en) * | 2000-03-15 | 2001-12-27 | Pennello Thomas J. | Method and apparatus for debugging programs in a distributed environment |
| US6941492B1 (en) * | 2002-02-05 | 2005-09-06 | Emc Corporation | Debugging tool for efficient switching between targets in a multi-processor environment |
| CN1499363A (en) * | 2002-10-24 | 2004-05-26 | 国际商业机器公司 | Method and device for setting break point on regulating integrated executable program of isomeric system structure |
Non-Patent Citations (4)
| Title |
|---|
| 一种并行程序性能分析工具的体系结构设计. 王忱，刘吟，蒋昌俊，张兆庆.信息技术与信息化，第5期. 2004 |
| 一种并行程序性能分析工具的体系结构设计. 王忱，刘吟，蒋昌俊，张兆庆.信息技术与信息化，第5期. 2004 * |
| 异构环境中的可视化和调试. A.贝戈林，J.佟伽拉.计算机工程与科学，第3期. 1994 |
| 异构环境中的可视化和调试. A.贝戈林，J.佟伽拉.计算机工程与科学，第3期. 1994 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1949185A (en) | 2007-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100435111C (en) | Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment | |
| CN102789413B (en) | A kind of debug system of concurrent program and method | |
| US7917894B2 (en) | Constraining source code and objects for analysis tools | |
| Balsamo et al. | Model-based performance prediction in software development: A survey | |
| Koschke et al. | An intermediate representation for integrating reverse engineering analyses | |
| US8756572B2 (en) | Debugger-set identifying breakpoints after coroutine yield points | |
| CN101727316B (en) | Parallel program integrated development system suitable for clustering system as well as application implementation method thereof | |
| US20080104096A1 (en) | Software development system | |
| JPH08202545A (en) | Object-oriented system and method for generation of target language code | |
| CN102622470A (en) | General car code conversion method | |
| CN1928816A (en) | Model drive for embedded system software and component development method | |
| CN101866315B (en) | Test method and system of software development tool | |
| CN103678135A (en) | System and method for achieving cross-process and cross-thread debugging in large-data environment | |
| CN101458633B (en) | Method for accessing host program by script program, and system and apparatus thereof | |
| Hamou-Lhadj et al. | A metamodel for the compact but lossless exchange of execution traces | |
| CN1627254A (en) | Method for debuging embedded system and equipment | |
| CN102968372B (en) | There is the program debugging system of process analysis function | |
| Mohapatra et al. | An overview of slicing techniques for object-oriented programs | |
| CN110109816A (en) | Test cases selection method and apparatus | |
| Tanter et al. | A versatile kernel for distributed AOP | |
| CN104317580A (en) | Integrated development environment for virtual test application (APP) | |
| Moore et al. | A scalable approach to MPI application performance analysis | |
| CN106681699A (en) | Online analyzing and calculating parallel programming auxiliary system and debugging method thereof for large power grid | |
| Wu et al. | Using graph patterns to extract scenarios | |
| CN112540907B (en) | Visual debugging method for parallel program on heterogeneous many-core |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081119 Termination date: 20111013 |