Medusa: The 300-Node Beowulf UWMLSC > Beowulf Systems > Medusa
Search the Medusa Web Pages
Links and Resources
- Frequently Asked Questions
- Bug Reporting
- CVS Archive
- UPS Status
- Ganglia Monitoring
- UWM LSC Group Computing
User Manual
- Backing Up Cron Jobs
- Becoming A User
- Connecting With Grid Tools
- LDAS Camp 2002
- LSC Grid Tools
- Obtaining LSC Grid Credentials
- Running Codes With Condor
Maintenance
- Adding A User
- Adding An LDAS User
- Bind Configuration
- Cloning A Node
- Data Recovery Procedure
- Documentation
- LDAS Administration
- Partitioning A Nodes Hard Disks
- Power Management
- Restoring A Database
- RPM Manual
- Scheduling LDAS DB Backup
- Slave Checklist
- Slave Work
Construction
- Benchmarking Results
- Benchmarking Software
- Channel Bonding
- Cluster-Dependencies
- Multimeter Instructions
- Node Prototype
- Paperwork
- Power Consumption
Design
- Bios Settings
- DHCP Configuration
- Fc3 Install
- File System
- Introduction
- IP Addresses
- Master Configuration
- Network Switch
- Physical Design
- Slave Configuration
- Software
Photo Gallery
- Hardware Pictures
- How To Take Pictures
- Setup Pictures
   
This document describes how to partition the HDD's used in our cluster.

As of this writing, you need to have an IBM xxxxxxx-1 drive as primary master
you need to have an IBM xxxxxxx-0 drive as secondary master, the cdrom as
secondary slave.  Optionally you may have a Fujitsu xxxxxxxx as primary slave.

The script is kept in CVS, under medusa/tools, and is called 
auto-partition.sh.  Run partition-disks.sh with no arguments for usage.  The script
does some basic checking for you, but assumes you know what you're doing.
Check this page for dead links, sloppy HTML, or a bad style sheet; or strip it for printing.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.