Minutes for the
ITC-Research Computing
Standing Committee
Meeting
June 26, 2006 at
10:30AM
(2015 Ivy Rd, 1st
floor Conference Room)
Members:
Alice, Bill, Dawn, Ed, Hamp, Kathy G., Katherine H., Jim, Mark S., Martha,
Nancy K., Robin, Sue Ellen, Steve, Terry, Tim S., Joe S., Tom S., Tim T.
Attending: Alice, Bill, Ed, Hamp, Kathy G.,
Katherine H., Jim, Mark S., Tim S., Joe S., Tom S., Tang (for Robin), Tim T.
Chair:
Tim T.
Recorder: Alice Howard
I. Corrections to minutes from last
Standing Committee meeting on April 24, 2006? No.
II.
Ongoing Discussion Topics:
*
“Condo” Cluster #2 – aka Dogwood: Timetable was to have it setup/running
in June, ready for users late July, early August. Update?
o
Starting
cabling and all the equipment we have is in racks – but are missing a Cisco
switch, it’s back-ordered– so will not be setup/running in June.
o
Any
update on SEAS faculty that wanted nodes or Infiniband cluster? None.
--> Jim will check with
SEAS.
o
Should
we order more nodes? Have space for 39 more nodes on the network switch and 31
empty spaces on the rack so could add 30 or so nodes fairly easily. However, Hamp thinks we’re at the edge
of cooling, power circuits and UPS now.
o
Have
new incoming EnviSci faculty (Stephan de Wekker) who wants 8 nodes now and
possibly more later.
o
New
release of ROCKS is imminent.
o
Not
sure if the partitioning of machine room in Carruthers is completed or not.
*
Cedar
Cluster: any problems/issues?
(Schedule of OS update in early August will be based on dogwood rollout)
o
None
*
Aspen
& Birch Clusters: (schedule of
OS update to Birch ONLY in early August will be based on dogwood rollout) Any
issues or concerns?
o
Agreed
at 4/06 meeting to keep Birch running/ under maintenance support another year.
o
Aspen
to be retired in late fall. Matt
Neurock may be interested in buying nodes from this. We just need it out of Carruthers. Need to make sure Matt
understands what he’s getting and associated headaches likely to come with it.
*
Oak
Cluster: From May meeting: interactive, round-robin login most
anytime. Will target this for
mid-June and Hamp will check with Chip about preferred dates.
o
Anyone
can login on any node now – will do the round-robin login around July 5. However we do NOT allow remote login
outside the UVa network on any research cluster – so one needs to login to
blue, etc. first or use VPN. Need
to make sure this is announced.
*
Teak
Cluster: Any issues or concerns?
o
Only
concern is age – but the problem is that this is the only system on which we
can legally run Gaussian/G03, which is used by the Chemistry department. So to discontinue support of Teak we
would need someone/Chemistry to get a license for Linux – or we would have to
migrate them to some other code.
*
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
Not
done yet.
*
Acquiring
monitoring tools as discussed in March, especially MPI Link Checker?
o
There
was some discussion about whether or not we really need this – especially if it
turns out to be expensive. Hamp
will renew his request for pricing, and then we can decide if it’s worth it.
o
Bill
P says there’s now commands built-in for the new Ethernet drivers that lets him
query their network communication settings to help debug problems. So probably don’t need MPI Link Checker
for this anymore. Hamp
thinks might be useful if getting InfiniBand cluster (and for Birch) since
communication problems are harder to detect and resolve. Tim asked Bill if he could let Ed
& Katherine know the new commands that are available.
*
Andrew
Grimshaw’s request for Linux front-end to campus grid project – waiting on next
version of AVAKI.
o
However
Marty Humphrey has almost finished a grid Windows client that does direct file
transfers and Katherine is experimenting with it.
*
Mitch
Rosen’s IT – Infrastructure Supporting Research Task Force. (https://www1.seas.virginia.edu/itrtf). (Maillist is itrtf at virginia dot edu) Early draft is circulating for input – no recent update or
“final” draft.
o
CI-TEAM
NSF proposal submitted on time with lots of input and assistance from Ed &
Katherine.
*
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 moving to Brown on 7/17 – the lab is now set up and will be called the
“Research Computing Lab”.
o
Alderman
move will happen over 8/7-8/9; this will mark move of rest of the Research
Computing Support Center out of Wilson.
Space in Alderman called the “Scholars’ Lab.”
o
The
main phone number (243-8800) is moving to Alderman; RCSC secondary number
(243-8799) moves to Research Computing Lab in Brown Library – staff phone
numbers stay the same.
o
Tim’s
office remains in Wilson.
III.
New Business
*
There
was discussion about removing telnet access from blue.unix later this summer
(and also ftp from everything except for anonymous ftp). After further discussion about how to
identify who is currently using telnet – and how to publicize this change, it
was decided to refer this issue to the Transitions Group.
*
Related
discussion about removing ability to ssh (secure shell / login) from OUTSIDE of
UVa domain to all research computing hardware. Hamp says we’re seeing numerous daily attacks. Would help
make us more secure.
We would need to announce that all research computing hardware will no
long be accessible unless inside UVa domain or running VPN.
*
Katherine
asked if we needed to get more PBS licenses in order to have enough for new
dogwood cluster. Bill thinks we’ve
got more than enough PBS licenses for the new cluster. Dawn A-M gets the ACHS one’s
separately.
Next
Scheduled Meeting of this Standing Committee group: Monday, August 28, 2006,
10:30 AM in ITC-2015 Ivy Road, Room #102.
Next
Scheduled Meeting of ITC ResComp Management team: Monday, July 24, 2006, 10:00
AM in ITC-2015 Ivy Road, Room #220.
Go to: ITC Research Computing Standing Committee Home Page