Minutes for the
ITC-Research Computing Management Meeting
(
Members: Alice, Hamp, Jim, Terry, Tim S., Tom S., Tim T.
Recorder: Alice Howard
For consideration and information – minutes
from last standing committee meeting on
II. Ongoing Discussion Topics:
· “Condo” Cluster #2 – aka Dogwood:
o
Update
on configuration and rollout:
still waiting on some cables and the next version of ROCKS – hoping they
both come this week. Looks like dogwood
could be ready by early August.
o
Katherine
and Ed have a “configuration” list for dogwood TimT
will email to the group in order to make sure we all know what’s going on
dogwood and reduce omissions and surprises.
o
Update
on SEAS faculty that wanted nodes or Infiniband cluster? Jim reported that Mitch
has the pricing and now the researchers are deciding – so the ball is in their
court now.
· Cedar
Cluster: any problems/issues? (schedule
of OS update will be based on dogwood rollout)
o
Hope
to schedule the OS update before school starts.
o
Avaki gateway setup – still waiting on new
software version release.
o
Could/should we move cedar to a 64-bit
cluster? Have had some requests. Could start testing on a mini 64-bit cluster
in the fall – but there is a lot of work to do.
Decided to wait and discuss this at next month’s meeting.
· Aspen
& Birch Clusters: (schedule of OS
update to Birch ONLY in early August will be based on dogwood rollout)
o
· Oak
Cluster: TimT will check with Hamp about whether or not
the round-robin login got enabled on July 5.
· Teak
Cluster: Any issues or concerns?
o
Lester
Andrews still asking Hamp about Gaussian/G03
for Linux – but we would need to sign a new agreement in order to get the Linux
version.
· Andrew Grimshaw’s request for
Linux front-end to campus grid project – waiting on next version of AVAKI
which has not been rewritten yet.
o
However Marty Humphrey has almost finished
a grid Windows client that does direct file transfers and Katherine is
experimenting with it. Hope to bring up
dogwood supporting Globus from the start for Marty’s client.
· Proposal for Linux cluster
support as ‘for-fee’ service. At
December meeting, Hamp said he’d translate current hourly Unix support rates
into table/information showing cost for supporting cluster of X size for Y
years.
o
TimT will check with Hamp.
· MPI Link Checker?
o
Hamp sent costs, seems too expensive and
not especially necessary unless we’re getting Infiniband cluster.
· Update on transition from clear
text passwords to Eservices for Unix logins/connections to blue.unix, HSM,
& Longtmp?
o
This transition will involve a major user
education issue – especially for blue.unix users.
o
No timeline for this project yet – it has
to wait until the password synchronization web site is up. (Note:
the ITC Transitions Group is handling the removal of telnet and ftp access
to blue.unix.)
· Mitch Rosen’s IT – Infrastructure
Supporting Research Task Force. (https://www1.seas.virginia.edu/itrtf). (Maillist is itrtf@virginia.edu) Final draft is
circulating for input. Katherine is
doing the final editing. Ed did most of
the appendices that detail comparisons to other institutions.
o
Mitch sent an update about our
participation in ORNL’s response to NSF RFP for next generation/iteration of
Supercomputing Centers: the Oak Ridge
Associated Universities core universities:
U Tennessee, Duke, Florida State, Georgia Tech, NC State, U. Va.,
Vanderbilt and Virginia Tech; and others including U Oklahoma and Rice have
agreed to team with Oak Ridge National Lab on a grant submission for NSF’s
Petascale supercomputer program. It is
listed as one award of $200 million dollars over a four year period, beginning
in 2010. The preliminary proposal is due
at NSF September 8. We are in touch with
the consortium schools, discussing potential roles and contributions from the
participating universities. He will
update us further as more details emerge.
· “Priority” access for
participants in condo clusters:
Katherine’s proposal for PBS “Service Units.”
o
There was some discussion about what
problem we are trying to solve. In
general condo users do not want their jobs to wait – but rather to run
immediately – and confirm that their purchase of nodes has given them an
advantage, or higher priority, etc.
o
Katherine proposes a model of 2-3 queues
where condo purchasers would have access to a higher priority/fast queue (and
maybe even an express queue).
o
Another model would be to actually reserve
nodes for condo purchasers – even if this meant that machines might be idle
sometimes.
o
TimT will talk with existing condo purchasers
to see how they respond to the “queue model” vs. the “own your own nodes”
model. We will also contact some schools
with successful condo cluster programs and see if we can learn anything from
their models and experience.
· Update on CSS Research Computing
support staff and services relocating to Brown Science & Engineering
Library and Alderman Library fourth floor in summer 2006.
o
Katherine
and Ed moved to Brown on 7/17. Called
the “Research Computing Lab” in Brown Library.
Will be fully functional by the start of classes, 8/23. RCSC secondary number (243-8799) moves to
Research Computing Lab in Brown Library.
o
Nancy
and rest of RCSC will move to Alderman over 8/7-8/9; this will mark move of
rest of the
o
Kathy
remaining in
o
Tim’s
office remains in
III. New Business
· Decided to continue these meetings monthly (as needed) and drop
the “all staff” alternating monthly meetings – ask Robin, Mark, Joe Simard, and
maybe Steve to join this management meeting – and ask staff to attend this
meeting as topics warrant.
Next Scheduled Meeting of ITC Rescomp
Management team: Monday, August 28, 2006, 10:30 AM in ITC-2015 Ivy Road, Room
#220.
Go to: ITC Research Computing Standing Committee Home Page