The Community for Technology Leaders
2010 19th International Conference on Parallel Architectures and Compilation Techniques (PACT) (2010)
Vienna, Austria
Sept. 11, 2010 to Sept. 15, 2010
ISBN: 978-1-5090-5032-1
pp: 99-109
John H. Kelm , University of Illinois at Urbana-Champaign, 61801, USA
Matthew R. Johnson , University of Illinois at Urbana-Champaign, 61801, USA
Steven S. Lumetta , University of Illinois at Urbana-Champaign, 61801, USA
Sanjay J. Patel , University of Illinois at Urbana-Champaign, 61801, USA
ABSTRACT
In this paper, we evaluate a set of coherence architectures in the context of a 1024-core chip multiprocessor (CMP) tailored to throughput-oriented parallel workloads. Based on our analysis, we develop and evaluate two techniques for scaling coherence to thousand-core CMPs. We find that a broadcast-based probe filtering scheme provides reasonable performance up to 128 cores for some benchmarks, but is not generally scalable. We propose a broadcast-collective network for accelerating probe filter misses, which extends scalability but falls short of supporting 1024 cores. We find that a sparse directory with an invalidate-on-evict policy can work well for many throughput-oriented workloads. However, the on-die structures required to achieve good performance carry a large performance and power overhead. To achieve thousand-core scalability with smaller and less associative sparse directories, we introduce WayPoint, a mechanism that increases directory associativity and capacity dynamically. Using less than 3% of total die area, WayPoint achieves performance within 4% of an infinitely large on-die directory.
INDEX TERMS
accelerator architecture, Cache coherence, probe filtering
CITATION
John H. Kelm, Matthew R. Johnson, Steven S. Lumetta, Sanjay J. Patel, "WayPoint: Scaling coherence to 1000-core architectures", 2010 19th International Conference on Parallel Architectures and Compilation Techniques (PACT), vol. 00, no. , pp. 99-109, 2010, doi:
99 ms
(Ver 3.3 (11022016))