Number of the records: 1
Microthreading as a Novel Method for Close Coupling of Custom Hardware Accelerators to SVP Processors
- 1.
SYSNO ASEP 0363714 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Microthreading as a Novel Method for Close Coupling of Custom Hardware Accelerators to SVP Processors Author(s) Sýkora, Jaroslav (UTIA-B)
Kafka, Leoš (UTIA-B)
Daněk, Martin (UTIA-B)
Kohout, Lukáš (UTIA-B) RIDNumber of authors 4 Source Title 2011 14th Euromicro Conference on Digital System Design Architectures, Methods and Tools DSD 2011. - Oulu, Finsko : IEEE Computer Society Conference Publishing Services, 2011 / Kitsos Paris - ISSN N - ISBN 978-0-7695-4494-6 Pages s. 525-532 Number of pages 8 s. Action 14th Euromicro Conference on Digital System Design Architectures, Methods and Tools DSD 2011 Event date 31.08.2011-02.09.2011 VEvent location Oulu Country FI - Finland Event type WRD Language eng - English Country FI - Finland Keywords microthreading ; SVP concurrency model ; UTLEON3 processor Subject RIV JC - Computer Hardware ; Software R&D Projects 7E08013 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) CEZ AV0Z10750506 - UTIA-B (2005-2011) Annotation We present a new low-level interfacing scheme for connecting custom accelerators to processors that tolerates latencies that usually occur when accessing hardware accelerators from software. The scheme is based on the Self-adaptive Virtual Processor (SVP) architecture and on the micro-threading concept. Our presentation is based on a sample implementation of the SVP architecture in an extended version of the LEON3 processor called UTLEON3. The SVP concurrency paradigm makes data dependencies explicit in the dynamic tree of threads. This enables a system to execute threads concurrently in different processor cores. Previous SVP work presumed the cores are homogeneous, for example an array of microthreaded processors sharing a dynamic pool of microthreads. In this work we propose a heterogeneous system of general-purpose processor cores and custom hardware accelerators. The accelerators dynamically pick families of threads from the pool and execute them concurrently. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2012
Number of the records: 1