 |
 |
Dear All, These meeting notes were posted on Confluence: https://confluence.slac.stanford.edu/display/hpsg/2014.03.20+--+Software+Meeting Best, Maurik ---------------------------------------- Agenda: Farm usage and Qweak request – All ECAL geometry update -- Holly ECAL clustering code update -- Kyle/Holly Alignment – Pelle Brief discussion: Documentation -- All Project updates? -- All Questions? -- All Notes: CPU use for the MDC: we expect that we need at least 400 cores continuously until May to finish the simulation and reconstruction. Maurik send an email to Harout Avakian to inform him of this. Disk Space - Stepan brought up that we need to formally ask for disk space. Sho stated that 30 TB of cache disk space should be good. We don't need that much work disk space. Maurik send an email to Harout to ask for 30 TB of cache disk and 3 to 5 TB of work disk. ECAL Geometry Update - Holly showed the updates for the ECAL geometry. She implemented the crystal support structure and some of the copper cooling pipes. She will implement a few additional pipes and some of the copper cooling plates. This will then be put into GDML for inclusion in the SLIC geometries. Clustering Algorithm - Holly and Kyle have updated the clustering algorithm EcalClusterIC. See her slides. The algorithm is ready for use and we encourage people to try it out. She still needs to look at and include the time information from the ECAL hits. This will help exclude noise hits from the cluster. It will also allow for an accurate cluster time. There was a request by Pelle to run this algorithm over the test run data and do a 1 to 1 comparison between this algorithm and the 3x3 cluster algorithm. This 3x3 algorithm was used for the test run analysis and it would be important to see if there are any differences in the results. Stepan requested some performance plots for the new algorithm: True energy versus cluster energy. Probably with some position dependence taken into account. Alignment - Pelle showed very promising results form the Millipede2 alignment code. See his slides. More work will be done to improve the procedure. It would be useful to have the tracker geometry reorganized so that it more closely reflects the physical layout of the tracker. This would make the alignment procedure more straight forward. HW Trigger code - Kyle has been working with Ben to update the simulation of the HW Trigger. He has updated the clustering to be closer to what the hardware does and is now looking at the trigger code, which has changed since the test run. He will give a more complete update later. Documentation - There was no outcry for better documentation. Jeremy did urge all coders to DOCUMENT YOUR CODE! (emphasis added), and please do so in a way that the comments show up correctly in the JavaDocs. Please note that you may also document other people's code if you think this is needed. ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the HPS-SOFTWARE list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1