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: 111-122
Daehoon Kim , Dept. of Computer Science, KAIST, Korea
Jeongseob Ahn , Dept. of Computer Science, KAIST, Korea
Jaehong Kim , Dept. of Computer Science, KAIST, Korea
Jaehyuk Huh , Dept. of Computer Science, KAIST, Korea
ABSTRACT
Although snoop-based coherence protocols provide fast cache-to-cache transfers with a simple and robust coherence mechanism, scaling the protocols has been difficult due to the overheads of broadcast snooping. In this paper, we propose a coherence filtering technique called subspace snooping, which stores the potential sharers of each memory page in the page table entry. By using the sharer information in the page table entry, coherence transactions for a page generate snoop requests only to the subset of nodes in the system (subspace). However, the coherence subspace of a page may evolve, as the phases of applications may change or the operating system may migrate threads to different nodes. To adjust subspaces dynamically, subspace snooping supports a shrinking mechanism, which removes obsolete nodes from subspaces. Subspace snooping can be integrated to any type of coherence protocols and network topologies. As subspace snooping guarantees that a subspace always contains the precise sharers of a page, it does not restrict the designs of coherence protocols and networks. We evaluate subspace snooping with Token Coherence on un-ordered mesh networks. For scientific and server applications on a 16-core system, subspace snooping reduces 44% of snoops on average.
INDEX TERMS
subspace snooping, cache coherence, snoop filtering
CITATION
Daehoon Kim, Jeongseob Ahn, Jaehong Kim, Jaehyuk Huh, "Subspace snooping: Filtering snoops with operating system support", 2010 19th International Conference on Parallel Architectures and Compilation Techniques (PACT), vol. 00, no. , pp. 111-122, 2010, doi:
312 ms
(Ver 3.3 (11022016))