From: Irwin Gaines To: Ruth Cc: gaines@dcdig.fnal.gov Subject: Status Reports and Manpower on Joint Projects Date: Wednesday, November 11, 1998 4:39 PM Run II Joint Project Physics Analysis Software Status report, 11/11/98 This project is just getting underway, based on the recent steering committee response to the PASREC committee recommendations. The project will have 3 main thrusts: 1) ROOT development and support 2) Evaluation, development and support of commercial tools 3) Collaboration with the LHC++ project ROOT: Current activities are minimal support, consisting of putting latest versions into kits from the cvs repository and performing testing. Starting on actual ROOT development awaits the resolution of licensing and collaboration issues; thjere are several potential development projects (as reviewed during Rene's last visit) that could start quickly once the collaboration agreements are in place and manpower is identified. Some evaluation from within the collaborations os also occuring. We need to start setting up a formal mechanism for the collaborations to make official requests for needed immprovements or bug fixes. Personnel available so far are Philippe (at a high duty fraction for some limited period of time), with help from Rob Kennedy and Pasha Murat. Long term we will likely need 3 FTE's (it is liikely thayt one of these will be a guest scientist; hopefully part of another might come from the new PAT hire). Commercial tools: We are so far concentrating on IDL and MATLAB. Jeff Kallenbach reports the current status: a) IDL I have a working prototype of a program which allows the user to sit at the interactive IDL session, and extract and plot variables from an HBook Ntuple using the variable name. It uses IDL "externals" (actually dso's), linked with the ZOOM Heptuple code, using KCC and CERN IRIX+5. This was really our first major milestone, and I will be meeting with Irwin and Patty soon to plan the next steps. b) MATLAB The MATLAB version is stuck with a compiler problem. I have received a suggestion from MathWorks, but haven't tried it yet because of the IDL progress. However, I am confident that we will be able to do the same sort of prototype with MATLAB that we have working with IDL. I should know more by the end of the week. c) Others I wouldn't mind giving a cursory look at some other products (Mathematica, SigmaPlot,...), just to be complete, but I think between the two of these we probably will be able to service most of our users. Between this, the anticipated expansions to both of these prototypes and vis, I could use some help, if someone wants to volunteer ;-). Thanks. Next steps with IDL include porting the IDL scripts to another platform besides IRIX (probably NT) and preparation of physicist oriented scripts to make it even easier to do the most usual types of physics applications. This work is proceeding very rapidly, and we will probably want to have a demo or computing techniques seminar soon. Personnel available so far are basically Jeff, with some technical help from others in PAT and with physicists input from Irwin, Rob, and others. We hope to do some D0 pilot projects with Mike Diesburg after the D0 installation is ready. Long term needs are roughly 1.5 FTEs (less if there is little physicist interest). LHC++: We will prepare an evaluation report on the current release of LHC++, with an eye towards more formal collaboration in the future. Among the areas to be studied are ease of setup and installation, licensing issues, portability (including desktop requirements for use), ease of use, backwards compatibility (ability to use n-tuples, both row wise and column wise), ability to save private persistent data, ability to share data files and analysis sessions (equivalent of last.kumac), etc. The report should be complete in of order one month. Personnel available so far are Simone Rolla from Tufts/CDF, with some potential contribution from PAT. Long term we probably need 0.5 FTEs for this part of the effort.