Speaker
Maarten Van Veghel
(Nikhef National institute for subatomic physics (NL))
Description
With the high demands on throughput of real-time data processing, in some cases, even existing fast ML inference libraries are not fast enough. Currently, hard-coded solutions native to the LHCb event model and algorithmic structure still win, both at GPU and CPU level. The goal is to develop solutions that have all these benefits but scale better and reduce the maintenance burden by for example using just-in-time compilation while still being able to seamlessly integrate in the event model, similar to the existing throughput-oriented (ThOr) functor infrastructure.
CERN group/ Experiment
LHCb
| Working area | Area 2: Optimal AI deployment for Online Data Processing |
|---|---|
| Project goals | Improve fast inference solutions at LHCb by converting existing fast ML solutions at both GPU and CPU level from hard-coded to just-in-time compilation while still having seamless integration with the event model and algorithmic structure. Improve scaling and maintainability |
| Timeline | 1 year |
| Available person power | 0.1 FTE |
| Additional person power request | 1 FTE |
| Is this an already ongoing activity? | No |
| Indicative hardware resources needs | Existing LHCb resources are sufficient |
Author
Maarten Van Veghel
(Nikhef National institute for subatomic physics (NL))