Difference between revisions of "UKI-NORTHGRID-LANCS-HEP"

From GridPP Wiki
Jump to: navigation, search
 
(No difference)

Latest revision as of 14:15, 25 July 2012

UKI-NORTHGRID-LANCS-HEP

Topic: HEPSPEC06


UKI-NORTHGRID-LANCS-HEP
OS+cores,Bits OS Kernel 32/64 mem gcc Total Per Core
Dual Core Intel 3.00 GHz Xeon SL4.6 2.6.9-78.0.1.ELsmp 64/32 2GB 3.4.6 10.86 5.43
Dual CPU, Quad core Intel Xeon E5440 @ 2.83GHz SL4.6 2.6.9-78.0.1.ELsmp 64/32 16GB 3.4.6 70.98 8.84
Dual CPU, Quad core Intel Xeon E5520 @ 2.27GHz SL4.6 2.6.9-78.0.1.ELsmp 64/32 24GB 3.4.6 97.56 12.20





Topic: Middleware_transition




gLite3.2/EMI


ARGUS - Not yet, need to plan this one out.

BDII_site - glite-BDII_site-3.2.10-1.sl5.x86_64

CE (CREAM/LCG) - glite-CREAM-3.2.10-0.sl5 / lcg-CE-3.1.40-0

glexec - Not yet (waiting on relocatable version)

SE - DPM-server-mysql-1.7.4-7sec.sl5.x86_64 (will upgrade to 1.8 towards the end of the month/start of October)

UI - SL5

WMS -NA

WN - glite-WN-3.2.9-0.sl5 tarball

Comments

We're planning to tentatively test the waters with UMD/EMI by installing a supplimentary cream CE with it, however like our peers we'd rather let the dust settle for a bit and only swap over when the benefits outweight the risks.

Topic: Protected_Site_networking


  • Both the local user and grid site are on separate, routable subnets
  • The Tier-2 is actually split across two, the "Grid" and the "HEC".
  • The Tier-2 is further split between a private and public VLANs.
  • There is a 10G backbone to the Tier 2 network.
  • There is a 1-Gb dedicated light path from the Tier-2 to RAL, and we also share the University's link to Janet (although I believe we are capped at 1Gb/s).
  • All switches are managed by ISS. We have access to the University Cacti pages.


Topic: Resiliency_and_Disaster_Planning



      • This section intentionally left blank


Topic: SL4_Survey_August_2011



lcg-CE: and having problems with CREAM, so will remain until that's done.

Topic: Site_information


Memory

1. Real physical memory per job slot:

2. Real memory limit beyond which a job is killed:

3. Virtual memory limit beyond which a job is killed:

4. Number of cores per WN:

Comments:

Network

1. WAN problems experienced in the last year:

2. Problems/issues seen with site networking:

3. Forward look:

Comments:

Topic: Site_status_and_plans



SL5 WNs

Current status (date): 22/3/2011
All nodes are on Sl5 or CentOs 5.

Planned upgrade:
NA

Comments:

SRM

Current status (date): 22/3/2011
All SL5 DPM 1.7.4-7

Planned upgrade:
Plan to upgrade to 1.8 in the near future.

Comments:

SCAS/glexec

Current status (date): 22/3/2011
Had some testing experiance, plan to roll it out after discussion with local admins and other work. Tarball install wanted, so will require liasing with other tar-sites.

Planned deployment:
ETA May

Comments:

CREAM CE

Current status (date): 22/3/2011
New cluster behind CREAM CE. Running without trouble.

Planned deployment:
Plan to deploy another cream ce for our other resources in April.

Comments: