From j.p.emerson@qmul.ac.uk Mon Jun 2 08:39:16 2003 Date: Sun, 1 Jun 2003 20:29:22 +0100 From: Jim Emerson To: Malcolm Stewart , Mike Irwin , Peredur Williams , Richard McMahon , Andy Lawrence Subject: WFAU WSA weekly meeting minutes, 30 May 2003 Please read these WFAU meeting minutes and feedback any comments to me. Jim -----Original Message----- From: Nigel Hambly [mailto:nch@roe.ac.uk] Sent: 30 May 2003 11:50 To: WFCAM Science Archive Team -- Clive Davenhall; Eckhard Sutorius; Harvey MacGillivray; Ian Bond; Mike Read; Nigel Hambly; Bob Mann Cc: Andy Adamson; Jim Emerson; Andrew Lawrence; Peredur Williams Subject: WFAU WSA weekly meeting minutes, 30 May 2003 Minutes of WFCAM Science Archive meeting: 30 May 2003 ----------------------------------------------------- ----------------------------------------------------- Present: NCH, ETWS, MAR, IAB Apologies: RGM, AL, HMG, JMS, PMW Actions discharged: ------------------- ACTION: IAB to do next iteration of SADD by the next meeting. ACTION: HMG to look into implications of the current WSA LAN design in C1 in case of any unforseen difficulties. ACTION: IAB to produce second iteration on SADD by Wed am (28th May). ACTION: ETWS (under guidance from NCH) to press forward with revision of WSA V1 schema scripts. - all done and discharged (thanks to all concerned!) Actions partly discharged but continuing: ----------------------------------------- ACTION: PMW to draft and action plan in the light of the review panel report. ACTION: ALL to review risk register and think of any new internal or external risks that should be documented in the risk register. ACTION: NCH to chase up Amos on the status of the final runs on the SSS data on IBM's bluedwarf machine. - continuing; thanks for progressing these. Actions carried forward from 23/05/03 meeting: ---------------------------------------------- ACTION: PMW to liaise with AL to make sure that Compute Support are made aware, from the highest "official" level, of our requirements, re. 100 Gbyte/day transfer from CASU, and to impress on them that the 1 Gbit/s FW should be their highest priority. - CONTINUES Specific points and new actions: -------------------------------- Project management: NCH again reported that he and PMW had worked all week on the VDFS continuation proposal (VDFS has now been absorbed into a larger road-map proposal called VEGA). Proposal development can be viewed at http://casse.ast.cam.ac.uk/twiki/bin/view/Community/VoVEGAVOProc Extra staff and hardware resources were being bid for the period Q2 2004 to Q3 2004, as well as continued resources to Q1 2007. Hardware: NCH reported that JNTD had been doing some exhaustive MemSpeed tests of the test server "eclipse". Ultra320 channel saturations tests indicate that 4 disks on one controller can yield 200 Mbyte/s; further tests on IBM hardware RAID adapter and alos software RAID 5 are ongoing. ACTION: NCH to update Twiki page with the latest benchmarks and tests. Networking: ETWS reported: "got the answer from Guy Rixon, that his server at Cambridge is running with an interim e-Science CA certificate. He will get e new host-certificate in a week or so. This may be caused by the circumstances that certificates issued by the Globus CA do not provide adequate security for deployment across an operational Grid at the national level. GridFTP here is using a Certificate from the UK e-Science Certification Authority. It is now tested for internal file transfers and worked well so the problem with Cambridge seems to be in the different certificates and namespaces therein." HMG reported: "I have looked into the hardware design for the suggested WSA Local Area Network and its consequences in room C1. I can foresee no major difficulties ,and indeed the system will be very straightforward. Computing support had suggested that we should isolate the WSA hardware from the site-wide LAN so that general on-site network performance will not be impacted by the heavy WFCAM data transfer load expected. This is very sensible. We already have made provision to purchase 2 LAN switches (each capable of 1Gb/s)in our bid for hardware, and 1 of these can readily be used for the job (the other will be used to fulfill local inter-connections within the "pixel" Beowulf). The only problem could be if the connection to the SRIF switch within C1 does not perform at 1 Gb/s. That being the case, we may need to purchase better cabling within C1 (at small cost) to improve the performance." Software: MAR reported: "Slow progress on parsing SQL queries results via Java servlet. I think I can see a way of incorporating the FITSWriter and AsciiFormatter classes I've developed into the existing SQLRetriever code." IAB reported much progress and third iteration of SADD: "The second iteration on the Software Architecture Design Document has been done. I'm now working on the third iteration. Wrapper modules and their interfaces are well defined now, as is how 15 of the curation use cases will import them. A lot of the components are well defined in that we can start coding them. First issue of this doc will probably contain around 50 pages." The meeting discussed how to close out the SADD and who should be asked to review it. Agreed that JMS and AB should be asked if they can look over the doc after completion of the 3rd iteration: ACTION: IAB to finalise issue 1 of the SADD by next meeting (6th June). ETWS has completed a revision of existing WSA scripts. The meeting discussed the next priority: either further script production based on current knowledge of UKIDSS surveys and their likely attributes, or generation of scripts for auto-documentation and analysis of the schema scripts. All agreed that the latter was probably more useful at this stage: ACTION: ETWS to start parsing scripts to generate HTML documentation from schema scripts, and analyse table parameters from the schema specifications. SSA: Production of intermediate files is still awaiting the final results from Amos' junk detection algorithms, and the new galaxy scale calibration in UKI. Likely that the latter will not be done in time for dump-out, so initial values will be transfered in the interim, and the calibration will have to updated in SQL Server (this will be a good test of analogous scenarios in the WSA anyway). The former is more of a hold-up since quality bits need updates on the basis of Amos' results. NCH will keep bugging Amos for news of progress. Miscellaneous: NCH reported attending a UKIDSS observing strategy meeting in London on the 28th; only issue relevant to WSA discussed was the question of standard field identifiers, which the other attendees seem happy to take on board. ACTION: NCH to type up and circulate these minutes DONM: 10am, Friday 6th June, plate library.