ScienceTools: (Jim) has no development news to report this week, but tag v9r16p1 finally got made yesterday.
From now on, instead of a weekly report, ScienceTools items may be found in (and added to) this blog. [Toby added the very first item shortly after the meeting.]
FSSC: (Richard) Any progress on the changes needed for FSSC to handle parallel ingest? (John) has heard nothing so far. (Jim) will implement once some (other) appropriate person has set out what needs to be done. (Richard) In the past that person has been Seth.
(Eric) has been working on port minus ROOT to Snow Leopard, but, discovering just now that ROOT 5.26 is in the new ST tag, he'll switch to working on that.
Pass 7 plans: (Anders) See particularly the roadmap. We will reprocess with Pass 7. (Richard) We probably will redo the preceding 1-1/2 years of data. (John) So this will lead to a redelivery at FSSC? (Eric) Timeframe? (Richard) Yes; expect it will take 4-6 weeks. What about parallel ingest at FSSC? (John) We don't have it yet. It doesn't require much work, but can't be done until there are some naming changes at the other end. He will send email to interested parties (Seth, Warren, and several more)
Documentation: (Chuck) sends the following:
All quiet on the WB front. I'm continuing work on the generic pipeline, and still need to get together with Tony and Dan regarding building two demo sites; one for the pipeline and one for the catalog.
Pass 7: (Richard) Tom had reprocessed up to October; now is caught up. The data are available from the Astro Server. In order to see where we are in this process (Anders) see the roadmap. By the way, the new Pass 7 will be Pass 7.3.
Pass 8: (Leon) Luca Baldini set out to finish the association of CAL clusters with ACD tiles, a worthy goal, and in the course of that decided it would be useful to calculate the 3rd moment (skewness) of the cluster to get insight into whether the track is going up or down. See his presentation for more details.
(Leon) has been working on technical upgrades:
(Tracy) is finishing off tree-based pattern recognition. It looks good for muons; now needs to test with gammas.
Still newer Studio (2010) is what Toby would like to see supported. Maybe a quicker route to this would be to update our CMT stuff to handle it. (Heather) is concerned about externals. Most of them now support our current target, Studio 2008 (aka 9.0) out of the box. We can download libraries rather than having to build them. [Official release date for Visual Studio 2010 Professional is April 12. ed.]
Externals (Heather)
GR v15r47p12gr11 This tag, intended for L1, has ROOT 5.26 so is compatible with the new ST tag. (Liz) Systests are in progress; she'll be looking them over presently. (Warren) will then test it in L1proc.
Mac news (Heather) Proposes that a mac currently running leopard be switched to snow leopard, and that the tiger machine be left alone since we are currently producing tiger builds.
(Richard) It costs about $200 to do the update. Once we decide finally on the machine to be upgraded, we need to ask unix-admin to do it.
RM database clean-up (Heather, Joanne) Kim has backed off on the original plan of automating clean up the dependency database since Navid had bad experiences with that. Instead he will add a new table which will keep track of which builds have been deleted. The installer can use the information in this new table in forming its list of builds available for download.
WIRED hang-up Direct from the source (Heather):
Tony only recently starting using a GR more recent than v15r24, and has run into problems trying to get Wired to run with something like GR v17r35p1gr02 (Pass 7 branch). Basically, when the second event is selected based on run/event id, Gleam crashes. I've been trying to narrow this down, and basically Wired uses a path through RootIo not used by Fred or typical RootIo..and that seems to be triggering some basic problem with switching between input files on the fly.
SCons topics
AOB (Warren) has been testing new merging code. Luca Baldini has verified fastmon and merit merging. Warren would like to do the same for digi and recon, but first he needs to recreate the files. He's hoping to use Luca's pyRoot script to diff the digi and recon trees.
previous | minutes index | next |