Systolic Processors as Second Level Triggers for High Energy Physics Experiments
نویسندگان
چکیده
This paper describes two fully programmable systolic processors for second level trigger processing in high energy physics experiments at CERN, Geneva, Switzerland. One of them has been used successfully since spring 1991 at the CERES/NA45 experiment for the recognition of Cherenkov rings in a ring image Cherenkov (RICH) detector. The processor consists of a 1m 2 systolic array of 176160, i.e. over 28,000 processing elements which are packed into 2220 VLSI chips that have been designed in 2 CMOS standard cell technology. The second processor, the ENABLE machine, is under design within the EAST/RD{11 collaboration at CERN that studies the data processing problems foreseen for the next generation of colliders like LHC, SSC, etc. It has been designed for the TRD detector of the LHC as a test application where a complex trigger decision has to be taken in less than 10 s. The ENABLE machine identiies arbitrary patterns in binary images by direct template matching. For each pattern the coincidences of the mask template and the picture under study are histogrammed. The machine is scalable in picture size as well as in the number of patterns processed concurrently. It exploits programmable gate array technology so that its architecture can be programmed into standard hardware. Both architectures and the consequences of their diierent implementation are discussed.
منابع مشابه
Systolic Processors in High Energy Physics*
Systolic processors are very well suited for the pattern recognition problems encountered in High Energy Physics. First, they provide an extremely high computing power due to their inherent massive parallelism. This computing power is needed for first and second level trigger processing tasks that have to be solved in the microsecond range and below. Second, they have a pipelined architecture t...
متن کاملA real-time application for the CS-2
of dividing the trigger process into several clearly separated steps, where each connected by a high—bandwidth switching network. Atlas follows the approach rates both the second-and third-level trigger. All processors and data buffers are The CMS solution uses a large general-purpose processor farm which incorpo slightly different approaches for implementing the second-level trigger system. Th...
متن کاملFast triggering in high-energy physics experiments using hardware neural networks
High-energy physics experiments require high-speed triggering systems capable of performing complex pattern recognition at rates of Megahertz to Gigahertz. Neural networks implemented in hardware have been the solution of choice for certain experiments. The neural triggering problem is presented here via a detailed look at the H1 level 2 trigger at the HERA accelerator, Hamburg, Germany, follow...
متن کاملUltra-Low-Energy DSP Processor Design for Many-Core Parallel Applications
Background and Objectives: Digital signal processors are widely used in energy constrained applications in which battery lifetime is a critical concern. Accordingly, designing ultra-low-energy processors is a major concern. In this work and in the first step, we propose a sub-threshold DSP processor. Methods: As our baseline architecture, we use a modified version of an existing ultra-low-power...
متن کاملGPUs for real-time processing in HEP trigger systems
We describe a pilot project (GAP GPU Application Project) for the use of GPUs (Graphics processing units) in online triggering applications for High Energy Physics experiments. Two major trends can be identified in the development of trigger and DAQ systems for particle physics experiments: the massive use of general-purpose commodity systems such as commercial multicore PC farms for data acqui...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2007