HLT2 and Sprucing sequences for pp reference run
Active lines
The included lines were presented at the PPG meeting on August 29:
A summary of the changes to the included lines:
- Removed prescales for charm cross section line
- Tuned prescales of QEE jet and dijet lines
- Added \gamma+j HLT2 and sprucing lines
- Added a lower-mass Drell-Yan line and loosened kinematic cuts and prescales on all Drell-Yan lines
- Loosened prescale on IFT isolated photon line
- Added IFT isolated photon sprucing line
Bandwidth estimates
Adds HLT2 and sprucing sequences for the 2024 pp reference run. The physics focus is on IFT, QEE (minus exotica), BandQ, and Charm (specifically production cross sections). Presented in the PPG here: https://indico.cern.ch/event/1446702/contributions/6107961/attachments/2917900/5120886/boettcher_PPG_240829.pdf
Bandwidth test is based on 1M MB events. Current bandwidth results WITHOUT SMOG2:
| Stream | BW |
|---|---|
| FULL | 0.63 GB/s |
| TURBO | 0.07 GB/s |
| TURCAL | 0.14 GB/s |
| Total, compressed | 0.84 GB/s |
| Total, uncompressed | 2.04 GB/s |
Note that this does not include the NoBias BW, as the lumi lines do not fire in MC. The target beam-beam NoBias rate for pp is 7.5 kHz in order to achieve an integrated luminosity of 50/nb, corresponding to about 600 MB/s for reconstruction and 400 MB/s for raw banks.
SMOG2 scaled to the expected set point (without minimum bias) contributes another ~150 MB/s spread across all streams.
Sprucing bandwidth without SMOG2:
| Stream | BW |
|---|---|
| B2CC | 0.2 MB/s |
| B&Q | 15.1 MB/s |
| IFT | 96.0 MB/s |
| QEE | 330.0 MB/s |
| SL | 24.4 MB/s |
| Total | 450.6 MB/s |
NoBias pp will contribute another ~1 GB/s, and SMOG2 (without minimum bias) will contribute another ~100 MB/s.
Data volumes
Offline computing has set a disk limit of 1.2 PB per replica. The data volumes were presented in the OPG meeting on October 3:
The expected conservative upper limit for data volumes:
- TURBO: 66 TB
- TURCAL: 124 TB
- FULL: 182 TB + 300 TB for SMOG2 minimum bias samples
- NOBIAS: 207 TB + 138 TB for raw banks
- Total: 1017 TB
TODO list
HLT2
-
Initial sequence -
Add lines (converted photon + jet, lepton + jet, dilepton + jet) -
Tune prescales/thresholds -
Bandwidth tests
Sprucing
-
Initial sequence -
Add lines -
Tune prescales/thresholds -
Bandwidth tests
Review checklist
Working groups
Please can each WG have at least one liason sign off. Expect to see a name and a link to a comment (can resolve immediately) summarising what was done for the review
-
QEE : @lugrazet : !3749 (comment 8561297) -
B2CC: @yimingli :!3749 (comment 8563976) -
IFT: @kmattiol: !3749 (comment 8561397) -
B&Q: @yajing: !3749 (comment 8561662) -
SLB_ @alupato: !3749 (comment 8570946) -
Charm @cacochat !3749 (comment 8570132)
RTA/DPA
-
@mvesteri as RTA WP3 coordinator : !3749 (comment 8516781) -
@abertoli for DPA