Analyzing PICL trace data with MEDEA

Conference · Fri Apr 01 00:00:00 EST 1994

OSTI ID:10137707

Merlo, A P ^[1]; Worley, P H ^[2]

---

1. Pavia Univ., (Italy). Dipt. Informatica e Sistemistica
2. Oak Ridge National Lab., TN (United States)

Execution traces and performance statistics can be collected for parallel applications on a variety of multiprocessor platforms by using the Portable Instrumented Communication Library (PICL). The static and dynamic performance characteristics of performance characteristics of performance data can be analyzed easily and effectively with the facilities provided within the MEasurements Description Evaluation and Analysis tool (MEDEA). A case study is then outlined that uses PICL and MEDEA to characterize the performance of a parallel benchmark code executed on different hardware platforms and using different parallel algorithms and communication protocols.

• Research Organization: Oak Ridge National Lab., TN (United States)
• Sponsoring Organization: USDOE, Washington, DC (United States); Consiglio Nazionale delle Ricerche, Rome (Italy); Ministero dell`Universita` e della Ricerca Scientifica e Tecnologica (Italy)
• DOE Contract Number: AC05-84OR21400
• OSTI ID: 10137707
• Report Number(s): CONF-9405115-1; ON: DE94009173
• Resource Relation: Conference: 7. international conference on modeling techniques and tools for computer performance evaluation,Vienna (Austria),4-6 May 1994; Other Information: PBD: [1994]
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
PARALLEL PROCESSING
M CODES
BENCHMARKS
COMPUTERS
DATA PROCESSING
COMPUTER CALCULATIONS
ALGORITHMS
990200
MATHEMATICS AND COMPUTERS