SPEED DMON: Cache Coherence on an Optical Multichannel Interconnect Architecture

This paper presents a low-overhead, high-performance cache coherence protocol and an interconnect architecture designed to exploit high-bandwidth point-to-point and broadcast features of an optical multichannel interconnect.SPEED DMONintegrates the virtues of snoopy-based schemes and directory-based schemes into a robust hybrid architecture. Directory-assist is used exclusively for decoupled read traffic to eliminate unnecessary broadcasts, while snoopy-assist is used exclusively for write and synchronization traffic to reduce directory overhead and synchronization complexities. The proposed scheme has the potential to reduce network latency by up to 85% as a result of its global independence between read and write operations, full concurrency in channel access, reduced link contention, and efficient broadcast of coherence and synchronization events.

[1]  J W Goodman,et al.  Design of an optical reconfigurable shared-bus-hypercube interconnect. , 1994, Applied optics.

[2]  T. M. Pinkston,et al.  Design considerations for optical interconnects in parallel computers , 1994, First International Workshop on Massively Parallel Processing Using Optical Interconnections.

[3]  P. Stenstrom.  A survey of cache coherence schemes for multiprocessors , 1990, Computer.

[4]  A.H. Gnauck,et al.  A transimpedance APD optical receiver operating at 10 Gb/s , 1992, IEEE Photonics Technology Letters.

[5]  Michel Dubois,et al.  Memory access buffering in multiprocessors , 1998, ISCA '98.

[6]  Paul R. Prucnal,et al.  Are crossbars really dead?: the case for optical multiprocessor interconnect systems , 1995, ISCA.

[7]  Bernard Glance,et al.  New advances on optical components needed for FDM optical networks , 1993 .

[8]  P. W. Dowd,et al.  Photonic architectures for distributed shared memory multiprocessors , 1994, First International Workshop on Massively Parallel Processing Using Optical Interconnections.

[9]  Paul Feautrier,et al.  A New Solution to Coherence Problems in Multicache Systems , 1978, IEEE Transactions on Computers.

[10]  Anoop Gupta,et al.  SPLASH: Stanford parallel applications for shared-memory , 1992, CARN.

[11]  Anant Agarwal,et al.  LimitLESS directories: A scalable cache coherence scheme , 1991, ASPLOS IV.

[12]  Robert J. Fowler,et al.  MINT: a front end for efficient simulation of shared-memory multiprocessors , 1994, Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[13]  Anoop Gupta,et al.  Reducing Memory and Traffic Requirements for Scalable Directory-Based Cache Coherence Schemes , 1990, ICPP.

[14]  Leslie Lamport,et al.  How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs , 2016, IEEE Transactions on Computers.

[15]  T M Pinkston,et al.  Smart-pixel-based network interface chip. , 1997, Applied optics.

[16]  Anoop Gupta,et al.  Memory consistency and event ordering in scalable shared-memory multiprocessors , 1990, [1990] Proceedings. The 17th Annual International Symposium on Computer Architecture.

[17]  N. A. Olsson,et al.  Surface-Emitting Microlasers for Photonic Switching and Interchip Connections , 1990 .

[18]  B. Mukherjee,et al.  WDM-based local lightwave networks. II. Multihop systems , 1992, IEEE Network.

[19]  B. Mukherjee,et al.  WDM-based local lightwave networks. I. Single-hop systems , 1992, IEEE Network.

[20]  R.H. Katz,et al.  A characterization of sharing in parallel programs and its application to coherency protocol evaluation , 1988, [1988] The 15th Annual International Symposium on Computer Architecture. Conference Proceedings.

[21]  M. Kuznetsov,et al.  Frequency tuning characteristics and WDM channel access of the semiconductor three-branch Y3-lasers , 1994, IEEE Photonics Technology Letters.

[22]  James K. Archibald,et al.  Cache coherence protocols: evaluation using a multiprocessor simulation model , 1986, TOCS.

[23]  A. Gupta,et al.  Correction to Cache Invalidation Patterns in Shared-Memory Multiprocessors , 1992 .

[24]  K. Kato,et al.  Packaging of large-scale integrated-optic N*N star couplers , 1993, IEEE Photonics Technology Letters.

[25]  Charles A. Brackett,et al.  Dense Wavelength Division Multiplexing Networks: Principles and Applications , 1990, IEEE J. Sel. Areas Commun..