Large-scale parallel computer system sectionalized parallel starting method

A computer system, large-scale technology, applied in computing, program loading/starting, program control design, etc., which can solve problems such as increased time, limited system availability, and long system startup time.

Inactive Publication Date: 2005-09-07
NAT UNIV OF DEFENSE TECH
View PDF0 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] When the system scale is very large, when this method selects the system BSP node and the system BSP node collects and builds the system resource information table, because it needs to communicate with all nodes in the system in turn, as the system scale increases (the increase in the number of nodes) , the time required for this serial process will increase linearly, so you will encounter the problem of too long system startup time, and the efficiency of system startup needs to be improved
At the same time, this method is inconvenient for system upgrade and reconfiguration, and the availability of the system is limited

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
  • Large-scale parallel computer system sectionalized parallel starting method
  • Large-scale parallel computer system sectionalized parallel starting method
  • Large-scale parallel computer system sectionalized parallel starting method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047]Figure 1 is a CC_NUMA system structure diagram. A computer system using the CC_NUMA structure is usually composed of many nodes connected to each other through an internal high-speed interconnection network. A node is the basic unit of the system. It has local CPU and memory, and can also have local I / O resources. Other nodes can remotely access these memory and I / O resources through the internal high-speed interconnection network. The nodes that make up the system include processing nodes and I / O nodes. Processing nodes are mainly responsible for computing tasks and run user applications on them. The IO node adopts the same structure as the processing node. In addition to participating in computing tasks, it also undertakes system input and output tasks. Each node adopts a distributed shared storage structure, and the memory under all nodes is uniformly addressed, and the cache consistency of the whole system is realized.

[0048] Fig. 2 is a block diagram of the int...

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

This invention discloses a large scale parallel computer system sectionalizing starting method for solving the technology problem of how to reducing starting time based on large scale of parallel computer system with CC_NUMA structure. Technology project is characterized by the following steps: dividing all nodes into groups, setting a sort of node controller to control them to choose BSP in parallel and by stages in Firmware and their initialization and allocation, including three steps of start namely choosing BSP, LBSP and GBSP. The invention has the advantages of reducing system starting time, improving system usability and getting conducive to expanding the system structure.

Description

technical field [0001] The invention relates to a starting method of a computer system, in particular to a parallel starting method of multiprocessors in a large-scale parallel computer system. Background technique [0002] CC_NUMA (Cache Coherent Non Uniform Memory Access Architecture) is a distributed shared storage system structure that implements the non-uniform storage access of the Cache consistency protocol. This structure is described in more detail in the white paper "ccNUMA Highly Scalable Server" published by SGI. A computer system using the CC_NUMA structure is usually composed of many nodes connected to each other through an internal high-speed interconnection network. A node is the basic unit of the system. It has local CPU and memory, and can also have local I / O resources. Other nodes can remotely access these memory and I / O resources through the internal high-speed interconnection network. Firmware (firmware) is the system software that controls the compute...

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): G06F9/445
CPCG06F9/4405
Inventor 蒋句平刘勇鹏刘光明田宝华苏湘玉屈晚霞郑明玲胡漪逸陈旭
Owner NAT UNIV OF DEFENSE TECH
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