In-Network Snoop Ordering (INSO): Snoopy coherence on unordered interconnects

Niket Agarwal, Li Shiuan Peh, Niraj K. Jha

Research output: Contribution to journalConference article

44 Scopus citations

Abstract

Realizing scalable cache coherence in the many-core era comes with a whole new set of constraints and opportunities. It is widely believed that multi-hop, unordered on-chip networks would be needed in many-core chip multiprocessors (CMPs) to provide scalable on-chip communication. However, providing ordering among coherence transactions on unordered interconnects is a challenge. Traditional approaches for tackling coherence either have to use ordered interconnects (snoopbased protocols) which lead to scalability problems, or rely on an ordering point (directory-based protocols) which adds indirection latency. In this paper, we propose In-Network Snoop Ordering (INSO), in which coherence requests from a snoop-based protocolare inserted into the interconnect fabric and the network orders the requests in a distributed manner, creating a global ordering among requests. Essentially, when coherence requests enter the network, they grab snoop-orders at the injection router before being broadcasted. A snoop-order specifies the global ordering of the particular request with respect to other requests. Before requests reach their destinations, they get ordered along the way, at intermediate routers and destination network interfaces. Our logical ordering scheme can be mapped onto any unordered interconnect. This enables a cache coherence protocol which exploits the low-latency nature of unordered interconnects without adding indirection to coherence transactions. Our full-system evaluations compare INSO against a directory protocol and a broadcast based Token Coherence protocol. INSO outperforms these protocols by up to 30% and 8.5%, respectively, on a wide range of scientific and emerging applications.

Original languageEnglish (US)
Article number4798238
Pages (from-to)67-78
Number of pages12
JournalProceedings - International Symposium on High-Performance Computer Architecture
DOIs
StatePublished - May 5 2009
Event2008 IEEE International Conference on Mechatronics and Automation, ICMA 2008 - Takamatsu, Japan
Duration: Aug 5 2008Aug 8 2008

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Fingerprint Dive into the research topics of 'In-Network Snoop Ordering (INSO): Snoopy coherence on unordered interconnects'. Together they form a unique fingerprint.

  • Cite this