An Abstract Interface for System Software on Large-Scale Clusters

Scalable management of distributed resources is one of the major challenges when building large-scale clusters for high-performance computing. This task includes transparent fault tolerance, efficient deployment of resources and support for all the needs of parallel applications: parallel I/O, deter...

Full description

Saved in:
Bibliographic Details
Published in:Computer journal 2006-07, Vol.49 (4), p.454-469
Main Author: Fernandez, J.
Format: Article
Language:English
Subjects:
Case studies
Communication
Computer networks
Computer peripherals
Debugging
Design
Distributed processing
Fault tolerance
High performance computing
Interfaces
Libraries
Network interface cards
Operating systems
Resource management
Scheduling
Software
Software engineering
Work stations
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Scalable management of distributed resources is one of the major challenges when building large-scale clusters for high-performance computing. This task includes transparent fault tolerance, efficient deployment of resources and support for all the needs of parallel applications: parallel I/O, deterministic behavior and responsiveness. These challenges may seem daunting with commodity hardware and operating systems, since they were not designed to support a global, single management view of a large-scale system. In this paper we propose and demonstrate an abstract network interface in the cluster interconnect to facilitate the implementation of a simple yet powerful global operating system. This system, which can be thought of as a coarse-grain SIMD operating system, can allow commodity clusters to grow to thousands of nodes, while still retaining the usability and performance of the single-node workstation. [PUBLICATION ABSTRACT]
ISSN:0010-4620
1460-2067
0010-4620
DOI:10.1093/comjnl/bxl020