Building high-performance clusters with OpenSSI

SPEED QUEEN

Article from Issue 84/2007
Author(s):

The OpenSSI framework rearranges processes for easy and transparent clustering.

A Single System Image (SSI) cluster is a collection of separate computers that appear as a single multi-processor system. SSI clustering platforms are typically implemented at the kernel level. In SSI high-performance clusters, the program does not even have to know it is running in a cluster. The only requirement is that the program spawn multiple processes. These processes then pass transparently among the computers in the cluster.

Buy this article as PDF

Download Article PDF now with Express Checkout
Price $2.95
(incl. VAT)

Buy Linux Magazine

SINGLE ISSUES
Print Issues
Digital Issues
SUBSCRIPTIONS
Print Subscriptions
Digital Subscriptions

Related content

  • Charly's Column

    Users log on to services such as SSH, ftp, SASL, POP3, IMAP, Apache htaccess, and many more using their names and passwords. These popular access mechanisms are a potential target for brute-force attacks. An attentive bouncer will keep dictionary attacks at bay.

    more »
  • HPC Cluster Basics

    The beginning for high-performance computing is understanding what you are trying to achieve, the assumptions you make to get there, and the resulting boundaries and limitations imposed on you and your HPC system.

    more »
  • Proxmox VE

    The Proxmox Virtual Environment has developed from an insider’s tip to a free VMware ESXi/ vSphere clone. We show you how to get started setting up a PVE high-availability cluster.

    more »
  • StarCluster

    Cloud computing has become a viable option for highperformance computing. In this article, we discuss the use case for cloud-based HPC, introduce the StarCluster toolkit, and show how to build a custom machine image for compute nodes.

    more »
  • Apache Cassandra

    The size and scope of today's Internet companies require more than your average SQL. Apache Cassandra is one of the NoSQL systems filling the need for high availability at scale.

    more »
comments powered by Disqus