Chapter 8. Job Scheduling with LPJS

Table of Contents

8.. Overview
8.. Using LPJS
LPJS Jargon
Cluster Status
Job Status
Using top
Job Submission
Terminating a Job
Terminating Stray Processes
File Sharing
Viewing Output of Active Jobs
Viewing Job Statistics
Job Sequences
Self-test

Before You Begin

Before reading this chapter, you should be familiar with basic Unix concepts (Chapter 3, Using Unix), the Unix shell (the section called “Command Line Interfaces (CLIs): Unix Shells”, redirection (the section called “Redirection and Pipes”), shell scripting (Chapter 4, Unix Shell Scripting) and job scheduling (Chapter 7, Job Scheduling.

Overview

LPJS, Lightweight, Portable Job Scheduler, is a minimalist resource manager and job scheduler for HPC clusters and HTC grids.

Unlike other resource managers, LPJS is designed to be portable to all POSIX operating systems, easy to install and configure, as easy to learn and use. The goal of LPJS is to remove as many barriers as possible to utilizing hardware for parallel computing.

Running a resource manager other than LPJS is a huge investment of time that is likely only worthwhile for large clusters or grids. LPJS, on the other hand, is simple enough for turning a few computers at home or in the office into a cluster or grid for small-scale parallel computing.