0. Status of production (Cosmic) e-Lab (10 mins)

   - general concerns
   - open problems

Bob: comments on status of cosmic
still concerns over reliability of cosmic
we have real users out there
Bob hears a growing level of frustration
sees uploads and data tailing off; less activity - sense that fall was not as active
feels that cosmic needs to focus on gettting the
single cosmic machine reliable (slow but always available)

more efforts on front end
metrics on useage
^ bob feels these two items are higher prio

1. 11:00 Code refactoring (15 mins)

Can Tibi, Ben, and Liz report on:

  a) what code state we are trying to get to, by when
   -- what intermediate steps, if these are aready clear
  b) what bodies of code are still involved
  c) what state they are in now
  d) who is doing what to integrate

  See http://www.ci.uchicago.edu/wiki/bin/view/I2U2/CleanupPlan20061219

see also bens email to the list
ben working with tibi on cosmic cms merge

ben sees no need to pull out the rls code even for cosmic
lot of effort for this
question on value of whether we remove RLS
mihael asks about alternative of removing it (need to take discusison offline)
tibi thinks we can build around it; mihael has concerns - take issue offline for email.

2. 11:20 Perl code improvements (15 mins)

  a) What problems are being addressed
  b) what improvements have been made
     - fixes
     - performance
      moving float to int; regex cleanups; rounding techniques; 
  c) where, how, who and when to integrate these
     - re how: shall we version the derived data?
     - pperl worth considering?

Need to deal with several file conversion efforts
need several convserions things - 
- split fix
- int conversion

Want to see real data so we know:
- what tr's are worth the most work
- what are the hot spots looking like

pperl uses a daemon

3. 11:40 Prep for Jan 24-26 meeting (10 mins)

  a) What are our training goals for the meeting?
  b) what needs to be done: how, who, when

Analysis tutorial - how best to do?
What methodology to suggest?

Need to resove what machines to do what (and how)
solve both code devl part and 

Figure who is going where - what sessions
Balance between what we do at meeting and what later

4. 11:50 Brief update on any progress in training (5 mins)

- TIbi is teating Jmeter, seems to be working for some initial things
- has converted Ben's initial test script into Jmeter
- Tibis notes are at PUTLINKHERE

-- Main.MichaelWilde - 04 Jan 2007
Topic revision: r4 - 2019-05-22, AdminUser
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback