Atlas-connect-l mailing list

[Rob Gardner <rwg AT hep.uchicago.edu>]

Folks,

Apologies, again, for the late notice but a few of us are not available today for various reasons.  We’ll endeavor to meet again next week.  The main things going on are:

1) Getting stampede connected.  Dave Lesny and Lincoln Bryant have been doing a lot of work setting up a Panda queue to submit production jobs into Stampede.  There has been much progress putting the puzzle pieces together, and its getting close.  I hope it will be fully functional by weeks end.  

2) Measuring the Parrot overhead - Suchandra has been working on this.  He should have a report for next week’s meeting.

3) Linking in BNL.  Given its strict security policies, we’re having to do special things to get those 1000s cores connected.  It should all be doable, perhaps by next week we’ll have it functional.

4) FAXbox usability.  A few issues have emerged creating load on faxbox.  This seems to be xrootd threads piling up.  More investigation into the cause is needed.  There is a new tutorial provided by David Champion (on the ConnectBook page) on using Faxbox with Globus Online, and we should have an http interface up today or tomorrow.  This will need some testing as well (volunteers?).

5) Tier 3 user adoption.  The UChicago ATLAS group continues to port their analysis to provide useful feedback. (Such as the faxbox load discovery above).  IU, UIUC, UCSC have been invited to start using it.   There does seem to be a psychological barrier to usage we need to understand.

I’ll send an announcement out by week’s end for next week’s meeting.

Cheers,

- Rob

---

Rob Gardner • Twitter: @rwg • Skype: rwg773 • g+: rob.rwg • +1 312-804-0859 • University of Chicago