Agenda for the ITC-Research Computing
Standing Committee Meeting
April 25, 2005 at 10:30AM in ITC-2015 Ivy Road, Room 102
Members: Alice, Bill, Dawn, Ed, Hamp, Kathy G., Katherine
H., Jim, Mark S.., Martha, Robin, Sue Ellen, Steve, Terry, Tim S., Joe S.,
Tom S.
Chair: Tim T.
Recorder: Alice Howard
I. Corrections to
minutes from last meeting on March 4, 2005 (Feb Mtg)?
II. Old Business:
- Cluster projects:
- "Condo" Cluster:
Will operate as 32-bit cluster for at least first year.
- Is 4 node test cluster setup? Would like to have Katherine start testing the "Fair Share" PBSPro setup we think will give researchers their 80% usage.
- Have four researchers participating: Leo Zhigilei (lz2n) - 8 nodes w/ ETF $; Jeff Shabanowitz/Dina Bai (Chemistry)- 8 nodes; Phil Parrish/Rich Gregory - 4 nodes; Matt Neurock - 4 nodes.
- see http://www.itc.virginia.edu/research/itc-clusters/itc-linux-cluster-purchase-dec04.pdf for details.
- Teak Cluster: Any outstanding issues?
- Oak Cluster: Any outstanding issues?
- Aspen & Birch cluster: Timeline for OS upgrade and upgrade to 2.6 kernel.
- /longtmp disks from Library available, thought possibly bring online mid-March?
- script to automate checking usage and reminding users.
- Linux license server, cluster of three lm1., lm2, & lm3.license.
Propose moving, if possible, all research software licenses off of aix.license onto linux.license server
- blue.unix upgrade of nodes to new 64-bit nodes and new version of OS. ITC-Transitions CDP is tracking as well. Is holmes converted over yet?
- Timeline/order of these projects?
- Discuss/ thoughts on Andrew Grimshaw's "grid clustering" presentation
III. New Business:
- Upgrades to PGI and Intel compilers on linux clusters. PGI to Version 6.0 and Intel to Version 8.1
- Scalapak doesn't work on Aspen and Birch. Need to install newest version of Intel compiler (8.1) in order for Scalapak to work. Have researchers who need.
- Need /jobtmp symbolic link on all clusters (aspen,birch, cedar, oak, teak, etc.)
- Jim Oberhauser's complaints about aspen.itc queue management and practice.
- Users allowed to get 18 nodes (36 CPUs) for parallel jobs and 24 nodes serial.
- Fair Share scheduling makes it possible for longer job to run before shorter job based on amount user's job have using cluster recently.
- VORTEX: Impact on most researchers, how to advertise/describe to researchers?
- Possibly have short article in May ResComp Newsletter?
V. Adjourn by 12:00PM and next meeting
- Next Scheduled Meeting of this group: Monday, June 27th, 10:30 AM in ITC-2015 Ivy Road, Room #102
- Next Scheduled Meeting of ITC ResComp Management team: Monday, May 23rd, 10:30 AM in ITC-2015 Ivy Road, Room #220
Suggestions, additions, corrections to: Tim
Tolson or Alice Howard

Go to: ITC Research
Computing Committee Home Page