Scalable input/output

Achieving system balance

Daniel A. Reed - Collection Scientific and engineering computation series

376 pages, parution le 08/04/2004

Ajouter à une liste

Indisponible

Résumé

As we enter the "decade of data," the disparity between the vast amount of data storage capacity (measurable in terabytes and petabytes) and the bandwidth available for accessing it has created an input/output bottleneck that is a major constraint on the effective use of scientific data for research. Scalable lnput/0utput is a summary of the major research results of the Scalable I/O Initiative, launched by Paul Messina, then Director of the Center for Advanced Computing Research at the California Institute of Technology, to explore software and algorithmic solutions to the I/O imbalance. The contributors explore techniques for I/O optimization, including: I/O characterization to understand application and system I/O patterns; system checkpointing strategies; collective I/O and parallel database support for scientific applications; parallel I/O libraries and strategies for file striping, prefetching, and write behind; compilation strategies for out-of-core data access; scheduling and shared virtual memory alternatives; network support for low-latency data transfer; and parallel I/O application programming interfaces.

Sommaire

I/O characterization and analysis
Collective I/O and large-scale data management
Building parallel database systems for multidimensional data
ADIO: a framework for high-performance, portable parallel I/O
Informed prefetching of collective input/output requests
Compiler support for out-of-core arrays on parallel machines
CLIP: a checkpointing tool for message passing parallel programs
Learning to classify parallel I/O access patterns
Thread scheduling for out-of-core applications with a memory server
A scalability study of shared virtual memory systems

Voir tout

Replier

Caractéristiques techniques

	PAPIER
Éditeur(s)	The MIT Press
Auteur(s)	Daniel A. Reed
Collection	Scientific and engineering computation series
Parution	08/04/2004
Nb. de pages	376
Format	17,6 x 22,8
Couverture	Broché
Poids	602g
Intérieur	Noir et Blanc
EAN13	9780262681421
ISBN13	978-0-262-68142-1