1. Quarterly report SouthernTier-2 Quarter 03 2005 P.D. Gronbech

2. September 2005Quarterly report: SouthGrid Current site status data SiteService nodes Worker nodes Local network connectivity Site connectivity SRMDays SFT failed Days in scheduled maintenance Security incidents this quarter which impact on Grid BirminghamSL304 LCG2.6.0 SL304 LCG2.6.0 100Mb/s1Gb/sNo320 BristolSL304 LCG2.6.0 SL304 LCG2.6.0 100Mb/s1Gb/sNo18190 CambridgeSL305 LCG2.6.0 SL305 LCG2.6.0 100Mb/s2.5Gb/sNo750 OxfordSL304 LCG2.6.0 SL304 LCG2.6.0 100Mb/s2.5Gb/sNo730 RAL PPDSL303 LCG2.6.0 SL304 LCG2.6.0 1Gb/s2Gb/sDcache300 1)Local network connectivity is that to the site SE 2)It is understood that SFT failures do not always result from site problems, but it is the best measure currently available. Results based on the old SFT page as this contains history, Site only deemed to have failed if it did not have a good set of results for a particular day.

3. September 2005Quarterly report: SouthGrid All GridPP Resources SitePromisedActual Integrated kSI2K hours until this quarter CPU (kSI2K) Storage (TB) Integrated kSI2K hours until this quarter CPU (kSI2K) Storage (TB) Birmingham17169601969.31952604222.9 *9.3 Bristol332880381.910424411.9 *1.9 Cambridge280320324.4350400404.4 Oxford206736023618.51042440119 **18.5 RAL PPD132276015111.885848098 ***5.8 Total572028065345.94308168491.839.9 1)The GridPP-Tier-2 MoUs made reference to integrated CPU over the 3 years of GridPP2. Under the “Promised – integrated kSI2K hours until this quarter” an estimate is provided of what the Tier-2 would have expected to provide to this quarter on the basis of planned installations. “Static kSI2K” shows what would currently be expected if all purchases planned to this quarter had been made and implemented. The actual columns show what has been delivered. 2)RAL PPD delayed purchase due to lack of use earlier in year. * The Bristol Babar cluster transferred to Birmingham ** Delayed purchasing due to lack of computer room *** Delayed purchasing due to earlier lack of use

4. September 2005Quarterly report: SouthGrid LCG resources SiteEstimated for LCGCurrently delivering to LCG Total job slots CPU (kSI2K) Storage (TB) Total jobs slots CPU (kSI2K) Storage (TB) Birmingham54 **58.5240281.92 Bristol2611.91.120.80.16 Cambridge4028.44.24439.61.97 Oxford174156.697466.63.2 * RAL PPD14415111.88271.756.4 Total ***438406.428.1242206.7511.73 1) The estimated figures are those that were projected for LCG planning purposes: http://lcg-computing-fabric.web.cern.ch/LCG-Computing-Fabric/GDB_resource_infos/Summary_Institutes_2004_2005_v11.htm 2) Current total job slots are those reported by EGEE/LCG gstat page. * This figure includes the older se also, not currently reported on the gstat pages. ** 50% of ATLAS Farm available to LCG. *** The total estimated above comes from the MOU spreadsheet, the total from the LCG projected planning spreadsheet was 200 KSI2K and 7TB Shortfall is due to delayed purchasing at RAL and Oxford

5. September 2005Quarterly report: SouthGrid VOs supported by site SiteALICEATLASBABARBIOMEDCDFCMSDTEAMDZEROHONEILCLHCBNA48PHENOSIXTZEUSTotal Birmingham1111011010100109 Bristol1100011000100016 Cambridge1101011000110007 Oxford11011110001011110 RAL PPD11110111111010112 Total552415512151223 0 => not supported 1 => supported

6. September 2005Quarterly report: SouthGrid Resources used per VO over quarter (KSI2K hours) Site CPUALICEATLASBABARBIOMEDCMSDTEAMDZEROHONEILCLHCBPHENOZEUSTotal Birmingham01211876890612083002885065090021535 Bristol02600160007900112 CambridgeN/A Oxford0140362702093317227000215780170260248 RAL PPD0235122373553547287312084970180781101321291088 Total03769525881351931021835501138204624411014914172983 1)Information currently available from APEL http://goc.grid-support.ac.uk/gridsite/accounting/tree/gridpp_view.phphttp://goc.grid-support.ac.uk/gridsite/accounting/tree/gridpp_view.php - please note these pages are still under development! Nb. This could be automated with an SQL/R-GMA query

7. September 2005Quarterly report: SouthGrid Usage by VO for Tier-2 Jobs July 2005Aug 2005Sep 2005 alice042 Atlas1602455494879 babar250020486509 Biomed170025511689 cms572191527 dteam472643346600 dzero745 hone720389705 ilc020 lhcb175228206995 Na48000 pheno09337 zeus28096331939 Data taken from goc accounting pages NormSumCPU (CPU hours normalised to 1KSPECint2000 CPU July 2005Aug 2005Sep 2005 alice000 Atlas9185914719363 babar6562497714342 Biomed1307521728390 cms691612262076 dteam2905411 dzero000 hone429830374047 ilc000 lhcb2084346529508 Na48000 pheno001101 zeus29206231371

8. September 2005Quarterly report: SouthGrid Storage resources in use per VO (TB) Site StorageALICEATLASBABARBIOMEDCMSDTEAMDZEROHONEILCLHCBNA48PHENOZEUSTotal Birmingham00.5980000.000000600.0027000000.6007006 Bristol000000.00000040000000 Cambridge00.24100.00020600.00000090000.0000040000.241214 Oxford00.09400.001100.00000210000.0000011000.0003020.095405 RAL PPD00.07560.1570.000420500.00126800.1000600.000032800.0007070.0007280.335816 Total01.00860.1570.001726500.00127200.1027600.000037900.0007070.001031.273136 Difficult to provide this for the period but we can at least show *current* usage. Numbers need to be provided by site Admins (> du – sh) but this will change under dCache.

9. September 2005Quarterly report: SouthGrid Usage by VO (CPU)

10. September 2005Quarterly report: SouthGrid Usage by VO (jobs) Nb: This can be extracted from APEL

11. September 2005Quarterly report: SouthGrid Progress over last quarter SiteSuccessesProblems/Issues BirminghamUpgraded to 2.6.0 Took part in Pre release testing of 2.6.0 Installed Babar clusters formerly at Bristol. BristolSite installed, connected on 5 th July and now maintained by Y. Coppens and P. Gronbech. Finalizing the HP funded post Recruited Sys admin on Rolling Grant Lack of Manpower has hindered progress CambridgeUpgraded to 2.6.0 Integrated LCG cluster in to Cam Grid Condor cluster. Still some ownership problems wrt to Condor users cf lcg pool uid’s. Apel Accounting does not yet support Condor OxfordUpgraded to SL 304 Installed LCG_2.6.0 New Sys Admin Started New Computer room is some way off and some nodes are over heating. SRIF funding has been secured to build it. The upgrade of the lcg cluster can commence as soon as the room is ready as funds are reserved. RALPPDUpgraded to 2.6.0 Install dcache

12. September 2005Quarterly report: SouthGrid Tier-2 risks General risks Lack of use by casual users. Feedback from jobs that go astray is not user friendly Mitigating actions Need better training for users and more useable software. Have Integrated UI’s with Local clusters for ease of use. Institute specific risks Lack of adequate computer room space at Oxford is still a problem. Several nodes are running hot. Slow progress on building of new computer room will delay upgrade of Oxfords resources. Use of Condor at Cambridge prevents monitoring statistics?? Bristol Manpower Lack of Bristol resources Mitigating actions Condor support to be added to APEL New sys admin due to start October 1 st. Local cluster will be used, SRIF funded eScience cluster to follow with luck

13. September 2005Quarterly report: SouthGrid Tier-2 planning for next quarter Setup and purchase integration test bed for SouthGrid use Coordinate use of this cluster within UK Testzone Install LCG-2.7.0 Install SRM at all sites Support some non LHC VO’s Investigate DPM at Birmingham, then install either dcache or DPM at other sites dependent upon results of testing. Possible new Hardware purchases. Start some inter site performance tests Prepare for SC4 involvement

14. September 2005Quarterly report: SouthGrid Objectives and deliverables for last quarter Objective/deliverableDue dateMetric/output Install LCG2.6.0 at all sitesLate July 2005Completed Testing of disk-to-disk transfers in preparation for Service Challenge 4 31 st September 2005Not yet started First sites to install SRM31 st September 2005RAL install dcache Birmingham started installing DPM Support non LHC vo’s31 st September 2005Birmingham, Cambridge, Oxford and RAL supported Biomed, and were largest UK T2 contributor.

15. September 2005Quarterly report: SouthGrid Objectives and deliverables for next quarter Objective/deliverableDue dateMetric/output Install LCG2.7.0 at all sitesLate October 2005 Testing of disk-to-disk transfers in preparation for Service Challenge 4 31 st December 2005 All sites to install SRM31 st December 2005 Continue support non LHC VO’s31 st December 2005

16. September 2005Quarterly report: SouthGrid Meetings, papers & effort Tier-2 coordinator effortComments 3 months AreaDescription Talks ConferencesGridpp 14 Birmingham Publications For Tier-2 coordinator:

17. September 2005Quarterly report: SouthGrid Summary & outlook South Grid technical meeting held in August continued to focus sites on rapid upgrades. All sites are running the latest release and will start installing SRM’s in preparation for SC4. There are continuing manpower issues at Bristol but this will ease shortly as the new Systems Administrator starts in October. The part HP funded post has also been finalised. Oxford will be able to expand resources once the new computer room is built. SRIF funding for this has been obtained for the room. Oxford new Sys Admin is in place, allowing the T2C more time to Coordinate! Yves Coppens is providing valuable help across SouthGrid. Bristol are now on line and it is hoped to expand their cluster once the local sys admin is in place.