Conveners
RDC5: Session #1 (Trigger and DAQ)
- Jinlong Zhang (Argonne National Laboratory)
- Zeynep Demiragli
RDC5: Session #2
- Zeynep Demiragli
- Jinlong Zhang (Argonne National Laboratory)
This paper presents the SLAC Neural Network Library (SNL), a specialized set of extensible libraries designed in High-Level Synthesis (HLS) for deploying machine learning structures on Field Programmable Gate Arrays (FPGAs), eFPGAs and ASICs. Positioned at the edge of the data chain, SNL aims to create a high-performance, low-latency FPGA implementation for AI inference engines. Utilizing the...
Machine learning based developments made for the level-1 trigger of the CMS experiment at LHC, both for Run-3 and HL-LHC eras will be presented. Unsupervised anomaly detection models are used in CICADA and AXOL1TL implementations using high-level synthesis on Xilinx Virtex-7 based boards for Run-3 running at full LHC clock rate digesting every bunch crossing within level-1 trigger latency...
Evolution of the ATLAS detector readout is driven by the rapid development of COTS network and computing systems. The Front-End Link eXchange (FELIX) system takes advantage of the new COTs components to reduce complexity and life-cycle effort. FELIX is an interface between the trigger and detector electronics and commodity switched networks for the ATLAS experiment at CERN. This rapid...
LuSEE-Night is a project to investigate the feasibility of measuring the fundamental physics processes occurring during the cosmic Dark Ages using instrumentation on the lunar surface. The "Dark Ages" refers to the cosmic era between the last scattering of the cosmic microwave background (CMB) and the time when the first stars and galaxies formed. Only cold, non-luminous hydrogen gas existed...
The use of High-Level Synthesis Languages (HLS) instead of VHDL or Verilog for FPGA code development is no longer a novelty. HLS allows for greater abstraction, enabling the handling of increasingly complex problems. The rapid prototyping and exploration of various ideas made possible by HLS would be respectively impossible and too time-consuming when using an FPGA hardware language. HLS has...
Machine learning is becoming ubiquitous across HEP. There is great potential to improve trigger and DAQ performance, and potentially in other real-time controls applications. However, the exploration of such techniques within the field in low latency/power FPGAs has just begun. We present hls4ml, a user-friendly software, based on High-Level Synthesis (HLS), designed to deploy network...
We present a description of a high-performance direct memory access (DMA) engine and kernel driver for data acquisition systems. The DMA engine is designed to support multiple incoming interleaved data channels simultaneously. The kernel driver enables multiple user-space clients to access the DMA engine for receiving or transmitting data, with the ability to create a memory map in the user...
Computing demands for large scientific experiments, such as the CMS experiment at CERN, will increase dramatically in the next decades.To complement the future performance increases of software running on CPUs, both in online TDAQ systems and offline data processing, explorations of coprocessor usage hold great potential and interest. We explore the novel approach of Services for Optimized...
The Inner Pixel Tracker (ITkPix) is the most important subdetector in ATLAS for tracking and vertexing of the charged particles produced in the collisions. Being closest to the beam pipe, it also has the highest flux of particles traversing through the material per unit area at any given time. During HL-LHC, the number of particle interactions in every bunch crossing will increase manifolds....
The muon campus program at Fermilab includes the Mu2e experiment that will search for a charged-lepton flavor violating processes where a negative muon converts into an electron in the field of an aluminum nucleus, improving by four orders of magnitude the search sensitivity reached so far.
The Trigger and Data Acquisition System (TDAQ) of the Mu2e projects consists of commercial,...
