Fault recovery on a massively parallel computer system to handle node failures without ending an executing job

A computer system, computing node technology, applied in the direction of non-redundant fault handling, computers, digital computer components, etc., can solve the increase of soft and hard faults, restart, parallel computer system inefficient hardware utilization of computer shutdown issues of time

Active Publication Date: 2009-11-25
INT BUSINESS MASCH CORP
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In massively parallel computer systems like Blue Gene, the problem of soft and hard failures increases significantly due to the complexity of the system and the number of computing nodes in the system
Furthermore, in the prior art, a failure in one node could cause an entire partition of the computer system to become unavailable, or result in the need to abort or restart jobs executing on the partition
[0006] Since computer system downtime and restarting jobs waste valuable system resources, without a more efficient way to recover from system failures caused by soft faults, the Parallel computer systems will continue to suffer from inefficient hardware utilization and unnecessary computer downtime

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Fault recovery on a massively parallel computer system to handle node failures without ending an executing job
  • Fault recovery on a massively parallel computer system to handle node failures without ending an executing job
  • Fault recovery on a massively parallel computer system to handle node failures without ending an executing job

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The present invention relates to an apparatus and method for recovering from a soft failure on a node of a parallel computer system without ending a job being executed on a node partition including the failed node. The preferred embodiment will be described in terms of the Blue Gene / L massively parallel computer developed by International Business Machines Corporation (IBM).

[0017] figure 1 A block diagram representing a massively parallel computer system 100 such as the BlueGene / L computer system is shown. The BlueGene / L system is a scalable system, where the maximum number of computing nodes is 65,536. Each node has an Application Specific Integrated Circuit (ASIC) 112 , also referred to as a bluegene / L computing chip 112 . The computing chip incorporates two processors or central processing units (CPUs) and is mounted on node daughter card 114 . The nodes typically have 512 megabytes of local memory. The node board 120 accommodates 32 node daughter cards 114 ea...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A method and apparatus for fault recovery of on a parallel computer system from a soft failure without endingan executing job on a partition of nodes. In preferred embodiments a failed hardware recovery mechanism on a service node uses a heartbeat monitor to determine when a node failure occurs. Where possible, the failed node is reset and re-loaded with software without ending the software job being executed by the partition containing the failed node.

Description

technical field [0001] The present invention relates generally to failure recovery on parallel computing systems, and more particularly to failure recovery on massively parallel supercomputers to handle node failures without ending executing jobs. Background technique [0002] Supercomputers are constantly being developed to handle complex computing jobs. These computers are especially useful to scientists working in high performance computing (HPC) applications including: life sciences, financial modeling, fluid dynamics, quantum chemistry, molecular dynamics, astronomy and space exploration, and meteorological modeling. Supercomputer developers have focused on massively parallel computer architectures to address this need for ever-increasingly complex computing demands. [0003] One such massively parallel computer being developed by International Business Machines Corporation (IBM) is the Blue Gene system. The Blue Gene system is a scalable system where the maximum numb...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F11/07G06F11/14
CPCG06F11/1441G06F11/0724G06F11/267G06F11/0793G06F1/24G06F13/00G06F11/16G06F15/00
Inventor D·达灵顿P·J·麦卡西A·彼得斯A·西德尼克
Owner INT BUSINESS MASCH CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap