Analyzing PICL trace data with MEDEA

Merlo, Alessandro P.; Worley, Patrick H

doi:10.1007/3-540-58021-2_25

Cited by 3 publications

(1 citation statement)

References 4 publications

(1 reference statement)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An advanced tool for these purposes is the MEasurements Description Evaluation and Analysis tool (MEDEA) [23], produced for CNR project "Sistemi Informatici e Calcolo Parallelo" , by the University of Pavia. Portable across systems supporting UNIX, X Windows and Motif, it has a graphical interface and supports the following stages of trace analysis: data manipulation, format manipulation, cluster analysis, fitting, functional description, data visualisation.…”

Section: Wat and Medeamentioning

confidence: 99%

Performance Analysis Tools in Europe

Pooley¹

1995

it - Information Technology

View full text Add to dashboard Cite

presents pointers to new techniques which are likely to be incorporated into usable tools in the near future. After a brief definition of the term performance the paper considers the techniques available to perform such an analysis and the tools which can be used to support such work. Tools are divided into queueing network tools, Petri net tools, process algebras and formal protocol languages and simulation tools. There is a brief look at alternatives to modelling and integrated environments. Under each of these headings a small number of the most representative and widely available tools are presented. Europäische Werkzeuge der Leistungsanalyse Dieser Beitrag wendet sich an Leser mit einem Interesse an derzeit verfügbaren Werkzeugen zur Leistungs-Modellierung von Rechensystemen und Netzwerken. Er enthält eine Übersicht über Werkzeuge, die seitens kommerzieller und akademischer Organisationen in Europa erhältlich sind. Da sich das behandelte Themenfeld in stürmischer Entwicklung befindet, sind zusätz-lich Hinweise auf neue Techniken enthalten, die voraussichtlich bald in derartigen Werkzeugen zugreifbar sein werden. Nach einer kurzen Definition des Begriffs Leistung werden Techniken betrachtet, welche der Leistungsanalyse zugrundeliegen. Die Werkzeuge werden eingeteilt nach ihrer Zugehörigkeit zu Warteschlangennetzen, Petri-Netzen, Prozeßalgebren und formalen Protokollspezifikationssprachen sowie Simulatoren. Kurze Abschnitte widmen sich Alternativen zur Modellierung und integrierten Bewertungsumgebungen.Unter den verschiedenen Stichworten wird jeweils eine begrenzte Anzahl der repräsentativsten und breit verfügba-ren Werkzeugen vorgestellt.

show abstract

Section: Wat and Medeamentioning

confidence: 99%

Performance Analysis Tools in Europe

Pooley¹

1995

it - Information Technology

View full text Add to dashboard Cite

show abstract

Workload Characterization Issues and Methodologies

Calzarossa

Massari

Tessera

2000

Performance Evaluation: Origins and Directions

View full text Add to dashboard Cite

Medea: a tool for workload characterization of parallel systems

Calzarossa

Massari

Merio

et al. 1995

IEEE Parallel Distrib. Technol., Syst. Appl.

View full text Add to dashboard Cite

Performance of parallel workload is heavily in uenced by h o w its own characteristics match the systems. Experimental approaches are particularly suitable, although di cult, for studying the behavior of the programs. The raw data collected by monitoring tools on real systems need to be processed for obtaining meaningful and easy{to{interpret information usable in identifying sources of poor performance. MEDEA (MEasurements Description, Evaluation and Analysis) is a software tool which meets such objectives. It has been designed as the natural extension of monitors in that it is typically used in strict conjunction with them. Various statistical and numerical techniques are applied to the collected data for deriving compact descriptions of the load, i.e., models, able to capture and reproduce its behavior. Visualization facilities are an integral part of our tool. As applications of MEDEA, a kernel and two real life programs (i.e., climate modelling and turbulent o w) running in di erent e n vironments are analyzed. Hints about possible code

show abstract

Analyzing PICL trace data with MEDEA

Cited by 3 publications

References 4 publications

Performance Analysis Tools in Europe

Performance Analysis Tools in Europe

Workload Characterization Issues and Methodologies

Medea: a tool for workload characterization of parallel systems

Contact Info

Product

Resources

About