检索结果-南通市图书馆

User-level failure detection and auto-recovery of parallel programs in HPC systems

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2021年第6期15卷 31-42页

作者： Guozhen ZHANG Yi LIU Hailong YANG Jun XU Depei QIAN State Key Laboratory of Software Development Environment Beijing 100191China Sino-German Joint Software Institute Beihang UniversityBeijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China Science and Technology on Space System Simulation Laboratory Beijing Simulation Center Beijing 100854China

As the mean-time-between-failures(MTBF)continues to decline with the increasing number of components on large-scale high performance computing(HPC)systems,program failures might occur during the execution period with high *** successful execution of the HPC programs has become an issue that the unprivileged users should be *** the user perspective,if the program failure cannot be detected and handled in time,it would waste resources and delay the progress of program ***,the unprivileged users are unable to perform program state checking due to execution control by the job management system as well as the limited ***,automated tools for supporting user-level failure detection and autorecovery of parallel programs in HPC systems are *** paper proposes an innovative method for the unprivileged user to achieve failure detection of job execution and automatic resubmission of failed *** state checker in our method is encapsulated as an independent job to reduce interference with the user *** addition,we propose a dual-checker mechanism to improve the robustness of our *** implement the proposed method as a tool named automatic re-launcher(ARL)and evaluate it on the Tianhe-2 *** results show that ARL can detect the execution failures effectively on Tianhe-2 *** addition,the communication and performance overhead caused by ARL is *** good scalability of ARL makes it applicable for large-scale HPC systems.

关键词： high performance computing parallel program failure detection failure auto-recovery

维普期刊数据库

Evolutionary Generation Approach of Test Data for Multiple Paths Coverage of Message-passing parallel programs

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2014年第2期23卷 291-296页

作者： TIAN Tian GONG Dunwei School of Information and Electrical Engineering China University of Mining and Technology

Test data generation, the premise of software testing, has attracted scholars in the software engineering community in recent years. Influenced by task partitioning, process scheduling, and network delays, parallel programs are executed in a non-deterministic way, which makes test data generation of parallel programs different from that of serial programs in essence. This paper investigated the problem of generating test data for multiple paths coverage of message-passing parallel programs. A mathematical model of the above problem was built based on each given path and its equivalent ones. It was solved by using a genetic algorithm to generate all desired data in one run. The proposed method was applied to five benchmark programs, and compared with the existing *** experimental results show that the proposed method greatly shortens the number of iterations and time consumption without reducing the coverage rate.

关键词： Software testing parallel program Multiple paths Test data Genetic algorithm

On the Problem of Optimizing parallel programs for Complex Memory Hierarchies

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 1994年第1期9卷 1-26页

作者：金国华陈福接 DepartmentofComputerScience ChangshaInstituteofTechnologyChangsha410073 DepartmentofComputerScience

Based on a thorough study of the relationship between array element accesses and loop indices of the nested loop, a method is presented with which the staggering relation and the compacting relation between the threads of the nested loop (either with a single linear function or with multiple linear functions) can be determined at compile-time,and accordingly the nested loop (either perfectly nested one or imperfectly nested one)can be restructured to avoid the thrashing problem. Due to its simplicity, our method can be efficiently implemented in any parallel compiler, and the improvement of the performance is significant as shown by the experimental results.

关键词： Optimization parallel program complex memory hierarchies SRIS,RSRIS compacted RSRIS

维普期刊数据库

A Method to Detect the Abnormal Phenomenonin PVM program Based on Petri Net

在线全文

学校读者我要写书评

暂无评论

Systems Science and Systems Engineering 1999年第2期9卷 165-178页

作者： JIANG Chanaiun(Department of Computer Science, Shandong Mining Institute, Tat’an 271019)(Department of Computer Science, Tough University, Shanghai 200092)LI Caiwei(Institute of Automation, Chinese Acadamy of Sciences, Beijing 100080)ZHANG Zhaoqing and QI

The parallel program verification can be divided into two aspects, the communicationand computation verification. In this paper, we mainly study the communication verification. First,we establish the Petri net specification for communication functions and basic statements of PVM;then, we give the steps of designing program verification model and abstract the abnormal phenomenonthat users often meet in the PVM parallel program into the characters of Petri net (including soleinformation, insufficient information, deadlock and livelock); finally, we give some detection algorithmto help users to eliminate these errors. In order to decrease the complexity of verification, we proposeseveral rules to simple the verification model.

关键词： parallel program verification Petri net verification model verification algorithm

维普期刊数据库