Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/76945
Title: DCache warn: an I-fetch policy to increase SMT efficiency
Authors: Cazorla Almeida, Francisco Javier
Ramírez Bellido, Alejandro
Valero Cortés, Mateo
Fernández García, Enrique 
UNESCO Clasification: 33 Ciencias tecnológicas
Keywords: Multi-threading
Cache storage
Instruction sets
Resource allocation
Multiprocessing systems
Issue Date: 2004
Publisher: Institute of Electrical and Electronics Engineers (IEEE) 
Conference: Proceedings - 18th International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM) 
Abstract: Simultaneous Multithreading (SMT) processors increase performance by executing instructions from multiple threads simultaneously. These threads share the processor's resources, but also compete for them. In this environment, a thread missing in the L2 cache may allocate a large number of resources for a long time, causing other threads to run much slower than they could. To prevent this problem we should know in advance if a thread is going to miss in the L2 cache. L1 misses are a clear indicator of a possible L2 miss. However, to stall a thread on every L1 miss is too severe, because not all L1 misses lead to an L2 miss, and this would cause an unnecessary stall and resource under-use. Also, to wait until an L2 miss is declared and squash the thread to free up the allocated resources is too expensive in terms of complexity and re-executed instructions. In this paper we propose a novel fetch policy, which we call DWarn. DWarn uses L1 misses as indicators of L2 misses, giving higher priority to threads with no outstanding L1 misses. DWarn acts on L1 misses, before L2 misses happen in a controlled manner to reduce resource under-use and to avoid harming a thread when LI misses do not lead to L2 misses. Our results show that DWarn outper-forms previously proposed policies, in both throughput and fairness, while requiring fewer resources and avoiding instruction re-execution.
URI: http://hdl.handle.net/10553/76945
ISBN: 0-7695-2132-0
DOI: 10.1109/IPDPS.2004.1303005
Source: Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2004 (Abstracts and CD-ROM)[ISBN 0769521320], v. 18, p. 1037-1046, 2004
Appears in Collections:Actas de congresos
Thumbnail
pdf
Adobe PDF (1,51 MB)
Show full item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.