Speaker
Description
Anomaly detection plays a key role as a novel strategy for trigger systems and real-time data analysis. This project, funded by Oracle Corporation and CERN openlab, focuses on developing and deploying AI models on the latest generation of AMD FPGAs (Versal with "adaptive intelligence" AI engine accelerators), for the L1 scouting system of CMS at the HL-LHC. The goal of this project is to implement a transformer-like model, that can be used for particle or event-level, classification, or fast-event reconstruction, at the level of the L1 Scouting hardware readout boards. Leveraging AI engines for the compute-heavy transformer architecture, in combination with the relaxed latency constraints of the L1 scouting (in comparison to the L1 itself), allows us to attempt more advanced ML algorithms/models than have previously been feasible in FPGAs.
CERN group/ Experiment
EP-CMD
| Working area | Area 2: Optimal AI deployment for Online Data Processing |
|---|---|
| Project goals | Intermediate goals (12m) Evaluate feasibility of implementation of transformer model on latest-and-greatest FPGA/SoC devices, and establish latency envelope. 18m Implementation running on hardware. Test and evaluate performance with simulated HL-LHC CMS physics data. ~30m Slice-test with other L1 trigger boards, system integration. |
| Timeline | Project started around 6 months ago, will continue for another 1.5-2.5 years in total |
| Available person power | One quest, ~10% of a staff. |
| Additional person power request | One quest/origin |
| Is this an already ongoing activity? | Yes |
| Indicative hardware resources needs | One AMD versal dev board with AI engines for in-lab use. Models range in price from 7-40kCHF. Some GPUs required for training and testing models. |