DocumentCode
233562
Title
Feign: In-Silico Laboratory for Researching I/O Strategies
Author
Luttgau, Jakob ; Kunkel, Julian M.
Author_Institution
Univ. Hamburg, Hamburg, Germany
fYear
2014
fDate
16-16 Nov. 2014
Firstpage
43
Lastpage
48
Abstract
Evaluating I/O performance of an application across different systems is a daunting task because it requires preparation of the software dependencies and required input data. Feign aims to be an extensible trace replay solution for parallel applications that supports arbitrary software and library layers. The tool abstracts and streamlines the replay process while allowing plug-ins to provide, manipulate and interpret trace data. Therewith, the application\´s behavior can be evaluated without potentially proprietary or confidential software and input data.Even more interesting is the potential of Feign as a virtual laboratory for I/O research: by manipulating trace data, experiments can be conducted; for example, it becomes possible to evaluate the benefit of optimization strategies. Since a plug-in could determine "future" activities, this enables us to develop optimal strategies as baselines for any run-time heuristics, but also eases testing of a developed strategy on many applications without modifying them.The paper proposes and evaluates a workflow to automatically apply optimization candidates to application traces and approximate potential performance gains. By using Feign\´s reporting facilities, an automatic optimization engine can then independently conduct experiments by feeding traces and strategies to compare the results.
Keywords
input-output programs; parallel processing; software performance evaluation; virtual instrumentation; Feign reporting facilities; I/O performance evaluation; I/O research; I/O strategies; application I/O; application behavior evaluation; application traces; arbitrary library layers; arbitrary software layers; automatic optimization engine; flexible-event imitation engine; in-silico laboratory; input data; optimal strategy development; optimization strategies; parallel applications; performance gain approximation; plug-ins; replay process abstraction; replay process streamlining; run-time heuristics; software dependencies; trace data interpretation; trace data manipulation; trace data provision; trace replay solution; virtual laboratory; workflow evaluation; Benchmark testing; Engines; Kernel; Laboratories; Optimization; Pipelines;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel Data Storage Workshop (PDSW), 2014 9th
Conference_Location
New Orleans, LA
Type
conf
DOI
10.1109/PDSW.2014.9
Filename
7016282
Link To Document