Stripe 82 data from winter'13 is public, so it's OK to distribute those for tests, etc. Re buildbot, is this the LSST buildbot, or you're running your own instance? -- Mario Juric, Sent from my Android device with K-9 Mail. Please excuse my brevity. Jacek Becla <[log in to unmask]> wrote: >Attendees: Andy H., Douglas, Fabrice, Daniel, Serge, Bill, Jacek > > >Integration cluster at IN2P3 > - have 50 nodes (part of the 300-node cluster) > until end of 2013, then will go away, too old > - IN2P3 wants to buy new small cluster for Qserv tests > - ~one rack would make sense, o(50) machines > - realistic spec that'd be useful (and relatively cheap): > - not many cores, eg two 4-core CPUs (8 cores total) > good enough > - drive per core, so 8 local drives > - we don't want any RAID > - 1TB drives, or 2TB if cheap enough > - 2GB RAM per core > >GLADE / EXTASCID > - (more at http://faculty.ucmerced.edu/__frusu/) > - interesting, but complicated > - not more parallel than Qserv > - might be useful to package lsst benchmark, including > queries and data and give to try to the GLADE developers > - we need to destroy scientific value of the data > before publishing (check with Mario?) > - schedule mtg with the key developer, 1-2 h > - might turn into some sort of collaboration > > >new xrootd client > - 70% complete > - can't try yet > - server side approaches stability > - ~abh/xssi/xrootd/source/xrd/ssi > - would be useful to look from api perspective now > - ready to try ~November > - not entirely clear if we should migrate when available, > or as part of the larger code rewriting > > >code refactoring / rewriting >---------------------------- > >very important to focus on tests > - automated test bed, with buildbot > - new metadata and new parser broke it > - some queries don't work > - work in progress to get that fixed now > - buildbot will rebuild everything, takes ~20 min now > - now buildbot runs on every commit to the master, > but no more than 1 per day > > - plus larger scale tests (daily? weekly?) > > > >proxy/lua > - would be cleaner if we didn't use it > - could try to link directly with mysql code > - the new connect engine not ready yet > > >Need to arrange a wiki trac account for Andy H, >he will be more involved in Qserv coding now, >not just XRootD > > >documents ideas, everyone should read ahead of mtg, >and come to design discussion prepared > > >logging > - want distributed logging with ability to query > - key/value flexibility would be very handy > - need to look from perspective how difficult it'd be > to migrate away if our tool of choice goes out > of fashion > >Start with: > - defining modules, interfaces > > >Draw a block diagram [Daniel] > - start with higher level, show to Bill or Andy, > drill deeper if needed > - need before hackathon (mid Nov) > > >Jacek > >######################################################################## >Use REPLY-ALL to reply to list > >To unsubscribe from the QSERV-L list, click the following link: >https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1 ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the QSERV-L list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1