ATExpert

doi:10.1006/jpdc.1993.1057

Journal of Parallel and Distributed Computing

Volume 18, Issue 2, June 1993, Pages 205-222

https://doi.org/10.1006/jpdc.1993.1057 Get rights and content

Abstract

ATExpert is a tool developed at Cray Research, Inc., which graphically displays the performance of parallel programs. Working behind ATExpert′s displays are several expert systems which simulate parallel execution and help a user gain insights into understanding the parallel performance of a program. ATExpert visually describes parallel performance at all levels in a program, guides the user to specific problem areas in the source code, explains why the problems are occurring, and suggests actions to resolve them.

References (0)

Cited by (23)

A theory and architecture for automating performance diagnosis
2001, Future Generation Computer Systems
This prospectus describes research to simplify programing of parallel computers. It focuses specifically on performance diagnosis, the process of finding and explaining sources of inefficiency in parallel programs. Considerable research already has been done to simplify performance diagnosis, but with mixed success. Two elements are missing from existing research:
- 1.
  There is no general theory of how expert programers do performance diagnosis. As a result, it is difficult for researchers to compare existing work or fit their work to programers. It is difficult for programers to locate products of existing research that meet their needs.
- 2.
  There is no automated, adaptable software to help programers do performance diagnosis. Existing software is either automated but limited to very specific circumstances, or in general, not automated for most tasks.
The research described here addresses both of these issues. The research will develop and validate a theory of performance diagnosis, based on general models on diagnostic problem-solving. It will design and evaluate a computer program (called Poirot) that employs the theory to automatically, adaptably support performance diagnosis.
Resource-aware meta-computing
2000, Advances in Computers
Meta-computing, is an increasingly popular and useful method of obtaining resources to solve large computational problems. However, meta-computer environments pose a number of unique challenges, many of which have yet to be addressed effectively. Among these are dynamicism in both applications and environments, and heterogeneity at several different levels. This chapter discusses current approaches to these problems, and uses them in the Active Harmony system as a running example. Harmony supports an interface that allows applications to export tuning alternatives to the higher-level system. By exposing different parameters that can be changed at runtime, applications can be automatically adapted to changes in their execution environment caused by other programs, the addition or deletion of nodes, or changes in the availability of resources like communication links. Applications expose not only options, but also expected resource utilization with each option and the effect that the option will have on the application's performance. We discuss how this flexibility can be used to tune the overall performance of a collection of applications in a system.
An experimental model to analyze OpenMP applications for system utilization
2011, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PerfExpert: An easy-to-use performance diagnosis tool for HPC applications
2010, 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2010
Integration of HIS in the performance-optimization field
2009, International Conference on Artificial Intelligence and Pattern Recognition 2009, AIPR 2009
A paradigm change: From performance monitoring to performance analysis
2009, Proceedings - Symposium on Computer Architecture and High Performance Computing

View all citing articles on Scopus

View full text

Journal of Parallel and Distributed Computing

Special Issue ArticleATExpert

Abstract

A theory and architecture for automating performance diagnosis

Resource-aware meta-computing

An experimental model to analyze OpenMP applications for system utilization

PerfExpert: An easy-to-use performance diagnosis tool for HPC applications

Integration of HIS in the performance-optimization field

A paradigm change: From performance monitoring to performance analysis

Special Issue Article
ATExpert