HEP-SCORE deployment TF meeting #24

Wednesday 2 Mar 2022, 16:00 → 17:00 Europe/Zurich

Zoomland

Helge Meinhard (CERN)

Minutes (approved) [PDF]

16:00 → 16:05 Welcome, note-taking, notes from previous meeting

Minutes (approved) of previous meeting [PDF]

16:05 → 16:15 Status of benchmark runs

Speaker: Gonzalo Menendez Borge (CERN)

2022-03-02_TF_Measurement_Collection_Update@TF.pdf

2022-03-02_TF_Measurement_Collection_Update@TF.pptx

16:15 → 16:25 Status of workloads

Speaker: Domenico Giordano (CERN)

memory-td805.png

- Alice
    - Received new set of scripts that use the new version of the Alice code
    - DB snapshot is on CVMFS
    - w.i.p. to adapt/implement the scripts in the HEP-workloads repo

- IGWN
    - Solved an issue in HEPscore causing failures when running the IGWN WL
    - A new version of HEPscore v1.4 will be released
    - w.i.p.: the IGWN WL will be finalised by Josh

- Juno and Belle2
     - single thread apps, fail in servers that have 128 cores (AMD EPYC 7452 32-Core Processor) ; RAM/core ~2 GB; and no swap.
    - When swap enabled the used memory is up to 2.7 GB/core ( see attached image)
    - Reported by Carles (PIC)

- Atlas Gen_sherpa
    - The cvmfs snapshotting of the image is huge (22GB), because of the unpacking of many not used files by the generator.
   - w.i.p.: A fix will be implemented

16:25 → 16:45 Update on analysis of data collected so far

Speaker: Randall Sobie (University of Victoria (CA))

Sobie-WLCG-TFBmk-Mar2.pdf

16:45 → 16:50 Any other business