Chapter 1. Introduction
Prev   Next

Chapter 1. Introduction

This document discusses the software that SGI provides for SGI GenomeCluster computations. This software, called CT-BLAST, is a sequence similarity search tool that supports database analysis performed on Linux clusters. CT-BLAST works with a scheduler and other tools to provide a completely integrated and customized bioinformatics computing solution. CT-BLAST is based on the Basic Local Alignment Search Tool (BLAST), developed by the National Center for Biotechnology Information (NCBI). This document explains how to perform administrative tasks in the SGI GenomeCluster environment.

Note: Examples in this document may not show the current RPM numbers.

Terminology

This document uses the following terminology:

  • Node. An SGI 1100, SGI 1200, SGI 1400, or SGI 1450 computer. A node is a host.

  • Execution node. A node on which a job is run.

  • Head node. The node that performs administrative tasks for the cluster and is where Portable Batch System (PBS) requests are submitted. Users log into the head node. A head node can also be an execution node, but there can be only one head node.

  • File server. An SGI Origin 200 server. The Origin 200 also acts as the batch server and scheduler for the cluster.

  • Cluster. One head node and multiple execution nodes.

  • Remote administration node. The node from which Lconsole administrative tasks are performed.

  • Serial multiplexer. A hardware and software product that allows a remote machine to perform administrative tasks. The Digi International EtherLite 16 is the default product available from SGI; the Comtrol RocketPort is an alternative product required for use with the SGIconsole cluster manager.

Assumptions

The instructions in this document assume the following:

  • You have experience with system administration tasks.

  • The software has already been installed.

  • You know the IP addresses and host names of each node in the cluster. (The host name is the output of the hostname (1) command.)

Contents of the SGI GenomeCluster Release

The SGI GenomeCluster release contains the following RPM packages:

  • sgi-cluster-tools

  • sgi-ctblast-docs

  • ctblast

  • ncbitoolkit (an open-source product shipped with SGI GenomeCluster)

  • data

SGI GenomeCluster also supports two third-party batch schedulers: LSF and PBS.

Space Requirements

To perform the installation, you must have sufficient amounts of free space in /tmp and /usr on each node in the cluster. For example, to install the software, you should have at least the following amounts of free space on each cluster node:

  • 42 MB in /tmp

  • 64 MB in /usr and /etc

  • 5 MB in /cache

Table 1-1 shows the amount of space required by each package.

Table 1-1. Approximate Space Requirements Per Package (in MB)

RPM Package

Head Node

Execution Node

sgi-cluster-tools

0.3

0.3

sgi-ctblast-docs

1.1

0

ctblast

5.9

0

ncbitoolkit

0

59.2

data

0

5.9

Ensure that you have enough disk space before installing SGI GenomeCluster.

Prev Table of Contents Next
About This Guide  Chapter 2. CT-BLAST