Meeting Notes for 10 October 2007 Refactored Code and By-Products Storage issues Email issue. (Bob and Tom J.) Tests (Pre and Post) SC2007 Systems (Argonne) Ligo CMS

Meeting Notes for 10 October 2007

Present: Liz, , Mihael, TomL, TomJ, Bob
Absent: Eric, Mike

Refactored Code and By-Products

Mihael using JMeter on www12. We will need to compare plots. [Mihael] It does compare plots. I've only converted one test so far.

Embedding analyses It will help Eric when he wants to run the analysis with SWIFT. We still have the issue of how Eric saves the plots from these analyses and gets them tagged with metadata so they are searchable in the e-Lab.

Tom J. said Nick D. worked on exposing the intermediate files in the Cosmic e-Lab. Tom wondered if Mihael could add this kind of functionality. (e.g., click here to see the intermediate files) Mihael can do this.

Analysis Queues Mihael added a button to the Data pages to see analyses in the queue. Currently the button only gives analyses from the current session. Eventually we need it to be for all jobs so students can go back to their results the next day.

Storage issues

When Mihael copied the files over from production so Bob could test, he removed threshold files from testing machine to make it possible to store so many files before we get the new machine up.

Email issue. (Bob and Tom J.)

How can we have an uptodate email list?

Need to have a reason for the users to want an email.

Currently no interface to update their information. (email). We would not want them to update their school, etc. because that could mess up the connection between data and school.

Currently use SQL Query to change email address.

Tests (Pre and Post)

Resetting Students in Research Groups to take tests in Cosmic e-Lab

Test sql commands on database on 12:9080

Mihael will send Liz the details for the database.

Improved Tests for Cosmic and new tests for CMS and Ligo Pre and Post testing. Tom L. wants to develop algorithm for producing questions. He will work with Jean on these.

SC2007

Tom L. will be getting out a better schedule. Confirmation # for Bob

Systems (Argonne)

Our project is perceived as a little erratic by Systems. Think we are making progress Maybe Mike needs to inform Rick that Mihael is the contact person for hardware. Eric will communicate with Mihael first.

Action Item: Mihael will ask Mike to send a message to Systems to clarify who is the contact so that systems does not get conflicting requests and will act on our requests. (DONE)

Ligo

Liz set up the software and clarified the process for entering references for Dale. Dale Ingram is starting to populate the references. Needs to do teacher pages. Liz provided templates.

CMS

Status of CMS e-Lab

What is the process to follow to come up with the best version.

Expose the code and the parameters.

Action item: Mihael will email Dan and set up a conversation.

Tom L. discussed the new focus to develop a full CMS e-Lab ? develop brand new one based on Monte-Carlo data and eventually real data. The test beam version will probably go away. Educators and scientists have to understand what are the fundamental questions that students can embrace.

Need to talk to Amelia people.

LHC and e-Lab fellows could be involved.

Action Item: Liz will send Tom J. the name of the LHC@FNAL fellow who discussed using data from CMS in Excel.

We can avoid worrying about updating the one study that does not work in the current CMS e-Lab. This will also mean that Dan K. may not have to spend a lot of time learning how to improve the version on www12:8080 that Tibi wrote, but turn his attention to Mihael?s methods. We could make a different link to the fourth study to go to the OGRE analysis at Notre Dame. This would allow students to do legitimate analyses of the study that has problems in the current version. It will require uploading the plots to include them in posters and will exclude these plots from the search.

Action item: Tom L. will look at this.

-- Main.LizQuigg - 11 Oct 2007