ABSTRACT
Tens and eventually hundreds of processing cores are projected to be integrated onto future microprocessors, making the global interconnect a key component to achieving scalable chip performance within a given power envelope. While CMOS-compatible nanophotonics has emerged as a leading candidate for replacing global wires beyond the 22nm timeframe, on-chip optical interconnect architectures proposed thus far are either limited in scalability or are dependent on comparatively slow electrical control networks.
In this paper, we present Phastlane, a hybrid electrical/optical routing network for future large scale, cache coherent multicore microprocessors. The heart of the Phastlane network is a low-latency optical crossbar that uses simple predecoded source routing to transmit cache-line-sized packets several hops in a single clock cycle under contentionless conditions. When contention exists, the router makes use of electrical buffers and, if necessary, a high speed drop signaling network. Overall, Phastlane achieve 2X better network performance than a state-of-the-art electrical baseline while consuming 80% less network power.
- V. R. Almeida, C. A. Barrios, R. R. Panepucci, M. Lipson, M. A. Foster, D. G. Ouzounov, and A. L. Gaeta. All-Optical Switching on a Silicon Chip. Optics Letters, 29(24):2867--2869, 2004.Google ScholarCross Ref
- J. Balfour and William Dally. Design Tradeoffs for Tiled CMP On-Chip Networks. In International Conference on Supercomputing, June 2008. Google ScholarDigital Library
- W. Bogaerts, P. Dumon, D. V. Thourhout, and R. Baets. Low-Loss, Low-Cross-Talk Crossings for Silicon-on-Insulator Nanophotonic Waveguides. Optics Letters, 32(19):2801--2803, 2007.Google Scholar
- G. Chen, Hui Chen, Mikhail Haurylau, Nicholas Nelson, Philippe Fauchet, Eby Friedman, and David Albonesi. Predictions of CMOS Compatible On-Chip Optical Interconnect. In Proceedings of the International Workshop on System Level Interconnect, April 2005. Google ScholarDigital Library
- W. Dally and B. Towles. Principles and Practices Of Interconnection Networks. Morgan Kaufmann, 2007. Google ScholarDigital Library
- N. E. Jerger, L.-S. Peh, and M. Lipasti. Virtual Circuit Tree Multicasting: A Case for On-Chip Hardware Multicast Support. In Proc. 35th IEEE/ACM International Symposium on Computer Architecture, June 2008. Google ScholarDigital Library
- A. Joshi, C. Batten, Y.-J. Kwon, S. Beamer, I. Shamim, K. Asanovic, and V. Stojanovic. Silicon-Photonic Clos Networks for Global On-Chip Communication. In Third ACM/IEEE International Symposium on Networks-on-Chip, May 2009. Google ScholarDigital Library
- N. Kirman, M. Kirman, R. K. Dokania, J. F. Martinez, A. B. Apsel, M. A. Watkins, and D. H. Albonesi. Leveraging Optical Technology in Future Bus-based Chip Multiprocessors. In Proc. IEEE/ACM 39th Annual International Symposium on Microarchitecture, Dec. 2006. Google ScholarDigital Library
- A. Kumar, Li-Shiuan Peh, Partha Kundu, and Niraj Jha. Express Virtual Channels: Towards the Ideal Interconnection Fabric. In Proc. 34th IEEE/ACM International Symposium on Computer Architecture, June 2007. Google ScholarDigital Library
- R. Kumar, Dean Tullsen, and Norman Jouppi. Core Architecture Optimization for Heterogeneous Chip Multiprocessors. In Parallel Architectures and Compilation Techniques, Sept. 2006. Google ScholarDigital Library
- N. McKeown. The iSLIP Scheduling Algorithm for Input-Queued Switches. ACM Transactions on Networking, 7(2):188--201, 1999. Google ScholarDigital Library
- D. Miller. Rationale and Challenges for Optical Interconnects to Electronic Chips. Proceedings of the IEEE, 88(6):728--749, 2000.Google ScholarCross Ref
- M. Paniccia, V. Krutul, R. Jones, O. Cohen, J. Bowers, A. Fang, and H. Park. A Hybrid Silicon Laser. White Paper, Intel Corporation, 2006.Google Scholar
- H. Park, Y. hao Kuo, A. W. Fang, R. Jones, O. Cohen, M. J. Paniccia, and J. E. Bowers. A Hybrid AlGaInAs-Silicon Evanescent Preamplifier and Photodetector. Optics Express, 15(21):13539--13546, 2007.Google Scholar
- K. Preston, P. Dong, B. Schmidt, and M. Lipson. High-Speed All-Optical Modulation Using Polycrystalline Silicon Microring Resonators. Applied Physics Letters, 92(15):151104, 2008.Google Scholar
- J. Renau, B. Fraguela, J. Tuck, W. Liu, M. Prvulovic, L. Ceze, S. Sarangi, P. Sack, K. Strauss, and P. Montesinos. Sesc simulator. http://sesc.sourceforge.net, 2005.Google Scholar
- Semiconductor Industry Association. International Technology Roadmap for Semiconductors, 2008.Google Scholar
- A. Shacham, Keren Bergman, and Luca Carloni. On the Design of a Photonic Network-on-chip. In First International Symposium on Networks-on-Chip, May 2007. Google ScholarDigital Library
- L. Shiuan Peh and William Dally. A Delay Model and Speculative Architecture for Pipelined Routers. In International Symposium on High-Performance Computer Architecture, Jan. 2001. Google ScholarDigital Library
- J. Tatum. VCSELs for 10 GB/s Optical Interconnects. Broadband Communications for the Internet Era Symposium Digest, 2001 IEEE Symposium on Emerging Technologies, Sept. 2001.Google ScholarCross Ref
- D. Vantrease, Robert Schreiber, Matteo Monchiero, Moray McLaren, Normal Jouppi, Marco Fiorentino, A. David, Nathan Binkert, Raymond Beausoleil, and Jung Ho Ahn. Corona: System Implications of Emerging Nanophotonic Technology. In Proc. 35th IEEE/ACM International Symposium on Computer Architecture, June 2008. Google ScholarDigital Library
- T. Woodward and A. Krishnamoorthy. 1-Gb/s Integrated Optical Detectors and Receivers in Commercial CMOS Technologies. IEEE Journal on Selected Topics in Quantum Electronics, 5(2):146--156, 1999.Google ScholarCross Ref
- F. Xia, L. Sekaric, and Y. Vlasov. Ultracompact Optical Buffers on a Silicon Chip. Nature Photonics, 1:65--71, 2006.Google ScholarCross Ref
- Q. Xu and M. Lipson. All-Optical Logic Based on Silicon Micro-Ring Resonators. Optics Express, 15(3):924--929, 2007.Google Scholar
Index Terms
- Phastlane: a rapid transit optical routing network
Recommendations
Firefly: illuminating future network-on-chip with nanophotonics
ISCA '09: Proceedings of the 36th annual international symposium on Computer architectureFuture many-core processors will require high-performance yet energy-efficient on-chip networks to provide a communication substrate for the increasing number of cores. Recent advances in silicon nanophotonics create new opportunities for on-chip ...
Phastlane: a rapid transit optical routing network
Tens and eventually hundreds of processing cores are projected to be integrated onto future microprocessors, making the global interconnect a key component to achieving scalable chip performance within a given power envelope. While CMOS-compatible ...
Iris: A hybrid nanophotonic network design for high-performance and low-power on-chip communication
On-chip communication, including short, often-multicast, latency-critical coherence and synchronization messages, and long, unicast, throughput-sensitive data transfers, limits the power efficiency and performance scalability of many-core chip-...
Comments