| Helm, B.R., A.D. Malony, and S.F. Fickas. Capturing and Automating Performance Diagnosis: the Poirot Approach. Proceedings of the Ninth 243 International Parallel Processing Symposium. 1995. Santa Barbara, CA: IEEE Computer Society Press |
....and a report for each failure. Some STEs with failure analysis capability use a test oracle, a subsystem that automatically analyzes software behavior and output during test execution. All purpose test oracles do not currently exist, but several domain specific oracles have been developed. Poirot [114] analyzes the execution of parallel programs to determine and isolate performance problems. TAOS GIL [115] compares a program s temporal specification against the trace of its implementation execution. TAOS Reactive [116] requires the tester to translate specification locations where certain ....
Helm, B.R., A.D. Malony, and S.F. Fickas. Capturing and Automating Performance Diagnosis: the Poirot Approach. Proceedings of the Ninth 243 International Parallel Processing Symposium. 1995. Santa Barbara, CA: IEEE Computer Society Press
....searched for. Finesse [16] directs the user to the overheads found in the application, and provides guidance for eliminating such overheads. Finesse focuses on shared memory programs, and its performance analyzer, which tries to find performance overhead, cannot be configured by the user. Poirot [10] is the design of a software tool which does not depend on any specific programming environment. To build such a design, Poirot s authors gathered several performance tools with the goal of formalizing performance bottlenecks of a parallel program. The search algorithm in Poirot uses a database ....
B. Robert Helm, Allen D. Malony, and Stephen F. Fickas. Capturing and automating performance diagnosis: The Poirot approach. In Proceedings of the 9th International Symposium on Parallel Processing (IPPS'95, pages 606--613, Los Alamitos, CA, USA, April 1995. IEEE Computer Society Press.
....search strategy, and finished finding bottlenecks 10 to 61 faster. In addition to improving search time, Deep Start often found more bottlenecks than the call graph search strategy. 1 Introduction Automated search is an effective strategy for finding application performance problems [7,10,13,14]. With an automated search tool, the user need not be a performance analysis expert to find application performance problems because the expertise is embodied in the tool. Automated search tools benefit from the use of structural information about the application under study such as its call graph ....
....automated performance diagnosis tools. The APART working group [3] provides a forum for discussing tools that automate some or all of the performance analysis process, including some that search through a problem space like Paradyn s Performance Consultant. For example, the Poirot approach [10] uses heuristic classification as a control strategy to guide an automated search for performance problems. Also, FINESSE [14] supports a form of search refinement across a sequence of application runs to provide performance diagnosis functionality. Search based automated performance diagnosis ....
Helm, B.R., Malony, A.D., Fickas, S.F.: Capturing and Automating Performance Diagnosis: the Poirot Approach. In Proc.
....Deep Start strategy found all bottlenecks in its search between 7 and 61 faster on average than the current strategy. Keywords: Automated performance diagnosis, search, sampling. 1 Introduction An automated search for performance problems is an effective strategy for performance diagnosis [7,10,14,15]. A user need not be an expert in performance analysis to be productive using an automated performance problem search tool; the analysis expertise is embodied in the tool s search strategy. The capability to automate searches for performance problems is enhanced by incorporating structural ....
....time required to find specific percentages of the known bottlenecks for each test application. The All Found rows indicate the average time required to find all bottlenecks that were found during a run. 5 22 01 17 space similar to Paradyn s Performance Consultant. For example, Helm et al. [10] describe the Poirot approach that uses heuristic classification as a control strategy to guide an automated search for performance problems. Also, the FINESSE project [15] supports a form of search refinement across a sequence of application runs to provide performance diagnosis functionality. 6 ....
B. R. Helm, A. D. Malony, and S. F. Fickas, "Capturing and Automating Performance Diagnosis: the Poirot Approach", Proceedings of the 1995 International Parallel Processing Symposium, 606--613, April 1995.
.... program run [27,54,55,61,69,77,80] In this common approach, much of the work is done manually by a knowledgable expert conducting the tuning study, frequently with the use of visualization tools [27,41,51,73] Recently, research has focused on automating the diagnostic process (steps 1, 2, and 3) [20,28,45,50,70]. Several projects [30,25,44,49] are attempting to redefine performance tuning as an interactive run time activity that may perform steering adjustments or debugging fixes as the application runs (step 4) Completing the spectrum, there is current research focused on complete automation of all ....
....problems found by estimating the reduction in execution time that would result from removing each bottleneck, and outputs a phase specific list of bottlenecks including location and cause. Several architectures have been proposed for a tool to perform automated diagnosis. The Poirot project [28] proposed an architecture for a tool to automatically diagnose parallel applications across a range of platforms. The project ended before a prototype was developed; however the work contributed an analysis of existing diagnostic approaches and an approach 12 to constructing a more general tool to ....
[Article contains additional citation context not shown here]
B. R. Helm, A.D. Malony, and S.F. Fickas. Capturing and automating performance diagnosis: the Poirot approach. Proceedings of the 1995 International Parallel Processing Symposium, pages 606--613, April 1995.
....to identify bottlenecks, decreases the amount of unhelpful instrumentation, and improves the usefulness of the information obtained from a diagnostic session. 1 INTRODUCTION Accurate performance diagnosis of parallel and distributed programs is a difficult and time consuming task. Recent research [1, 2, 14, 3, 4] examines possible approaches for automating, and thereby simplifying, the process of diagnosing a single program run. This paper describes how historical performance data, i.e. data gathered in one or more previous executions of an application, can be used to increase the effectiveness of ....
B. R. Helm, A. D. Malony, and S. F. Fickas. Capturing and automating performance diagnosis: the Poirot approach. In Proceedings, 1995 International Parallel Processing Symposium, pages 606--613, April 1995.
....machine. The most common motivation for developing high performing programs is that parallel machines are expensive resources that must be utilized to their maximum potential to justify their costs. However, the process of performance debugging (the iterative application of performance diagnosis [11] and tuning) invariably requires access to the target parallel platform, since the majority of the parallel performance tools are based on the measurement and analysis of actual program execution. As a consequence, during performance debugging, a parallel system is typically running less than ....
R. Helm, A. D. Malony and S. F. Fickas, Capturing and Automating Performance Diagnosis: The Poirot Approach, Proc. International Parallel Processing Symposium
....runtime tools can be built. A second direction is towards more expanded support for performance experimentation and diagnosis, particularly for scalability analysis. We are attempting to integrate a simulation based performance extrapolation tool and a semi automated performance diagnosis system [13] into the toolset. In addition, we are working on better ways to maintain a database of program measurements that record the progress of a sequence of experiments and to display comparative performance behaviors across multiple executions for different numbers of processors. Finally, we are ....
B. Robert Helm, Allen D. Malony, Steve Fickas, Capturing and Automating Performance Diagnosis: The Poirot Approach, to be published in Proc. 9th Inter. Parallel Processing Symp. (IPPS '95), Santa Barbara, California, April 1995.
....debugging tools, and modeling. 1 Introduction One of the foremost challenges for a parallel programmer is to achieve the best possible performance for an application on a parallel machine. For this purpose, the process of performance debugging (the iterative application of performance diagnosis [8] and tuning) is applied as an integral part of a parallel program development methodology. Application of performance debugging in practice has invariably required the development of performance tools based on the measurement and analysis of actual parallel program execution. Parallel performance ....
R. Helm, A. D. Malony and S. F. Fickas, Capturing and Automating Performance Diagnosis: The Poirot Approach, Proc. International Parallel Processing Symposium
....debugging tools, and modeling. 1 Introduction One of the foremost challenges for a parallel programmer is to achieve the best possible performance for an application on a parallel machine. For this purpose, the process of performance debugging (the iterative application of performance diagnosis [11] and tuning) is applied as an integral constituent part of a parallel program development methodology. Application of performance debugging in practice has invariably required the development of performance tools based on the measurement and analysis of actual parallel program execution. Parallel ....
R. Helm, A. D. Malony and S. F. Fickas, Capturing and Automating Performance Diagnosis: The Poirot Approach, Proc. International Parallel Processing Symposium
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC