Abstract
Conventional stream architectures focus on exploiting ILP and DLP in the applications, although stream model also exposes abundant TLP at kernel granularity. On the other side, with the development of model VLSI technology, increasing application demands and scalability challenges conventional stream architectures. In this paper, we present a novel Tiled Multi-Core Stream Architecture called TiSA. TiSA introduces the tile that consists of multiple stream cores as a new category of architectural resources, and designed an on-chip network to support stream transfer among tiles. In TiSA, multiple levels parallelisms are exploited on different granularity of processing elements. Besides hardware modules, this paper also discusses some other key issues of TiSA architecture, including programming model, various execution patterns and resource allocations. We then evaluate the hardware scalability of TiSA by scaling to 10s~1000s ALUs and estimating its area and delay cost. We also evaluate the software scalability of TiSA by simulating 6 stream applications and comparing sustained performance with other stream processors and general purpose processors, and different configuration of TiSA. A 256-ALU TiSA with 4 tile and 4 stream cores per tile is shown to be feasible with 45 nanometer technology, sustaining 100~350 GFLOP/s on most stream benchmarks and providing ~10x of speedup over a 16-ALU TiSA with a 5% degradation in area per ALU. The result shows that TiSA is a VLSI- and performance-efficient architecture for the billions-transistors era.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Khailany, B.: The VLSI Implementation and Evaluation of Area-and Energy-Efficient Streaming Media Processors. PhD thesis, Stanford,University (2003)
Rixner, S.: Stream Processor Architecture. Kluwer Academic Publishers, Boston (2001)
Bond, R.: High Performance DoD DSP Applications. In: 2003 Workshop on Streaming Systems (2003), http://catfish.csail.mit.edu/wss03/
Wen, M., Wu, N., Li, H., Zhang, C.: Multiple-Morphs Adaptive Stream Architecture. Journal of Computer Science and Technology 20(5) (September 2005)
Kozyrakis, C.E., et al.: Scalable Processors in the Billion-Transistors Era: IRAM. IEEE Computer 30(9) (September 1997)
Khailany, B., Dally, W.J., Kapasi, U.J., Mattson, P., et al.: Imagine: media processing with streams. IEEE Micro (March/April 2001)
Taylor, M.B., et al.: Evaluation of the Raw Microprocessor: An Exposed-Wire-Delay Architecture for ILP and Streams. In: ISCA 2004 (2004)
Sankaralingam, K., et al.: Exploiting ILP, TLP, and DLP with the Polymorphous TRIPS architecture. In: 30th Annual International Symposium on Computer Architecture (May 2003)
Hofstee, H.P.: Power Efficient Processor Architecture and the Cell Processor. In: Proc. of the 11th International Symposium on High Performance Computer Architecture (February 2005)
Fang, J.: Challenges and Opportunities on Multi-core Microprocessor. In: Srikanthan, T., Xue, J., Chang, C.-H. (eds.) ACSAC 2005. LNCS, vol. 3740, pp. 389–390. Springer, Heidelberg (2005)
Mattson, P.R.: A Programming System for the Imagine Media Processor. PhD thesis, Stanford University (2002)
Dally, W.J., et al.: Merrimac: Supercomputing with Streams. In: Proc. of Supercomputing 2003 (November 2003)
Wen, M., Wu, N., Zhang, C., Wu, W., Yang, Q., Xun, C.: FT64: Scientific Computing with Stream. In: Aluru, S., Parashar, M., Badrinath, R., Prasanna, V.K. (eds.) HiPC 2007. LNCS, vol. 4873, pp. 209–220. Springer, Heidelberg (2007)
Larus, J.: SPIM: A MIPS Simulator, http://pages.cs.wisc.edu/~larus/spim.html
Das, A., Mattson, P., Kapasi, U., Owens, J., Rixner, S., Jayasena, N.: Imagine Programming System User’s Guide 2.0 (June 2004), http://cva.stanford.edu/Imagine/project/
Mattson, P.: Communication scheduling. In: Proceedings of the Ninth International Conference on Architectural Support for Programming Languages and Operating Systems, Cambridge, MA (November 2000)
Nuwan, S.: Jayasena, Memory Hierarchy Design for Stream Computing. Stanford Ph.D. Thesis (2005)
Khailany, B., Dally, W.J., Rixner, S., Kapasi, U.J., Owens, J.D., Towles, B.: Exploring the VLSI Scalability of Stream Processors. In: Proceedings of the 9th Symposium on High Performance Computer Architecture, Anaheim, California (February 2003)
Erez, M.: Merrimac - High-Performance and High-Efficient Scientific Computing with Streams. PhD thesis, Stanford University (2006)
Das, A., Mattson, P., Kapasi, U., Owens, J., Rixner, S., Jayasena, N.: Imagine Programming System Developer’s Guide (2002), http://cva.stanford.edu/Imagine/project/
Mai, K., Paaske, T., Jayasena, N., Ho, R., Dally, W.J., Horowitz, M.: Smart memories: A modular reconfigurable architecture. In: International Symposium on Computer Architecture (June 2000)
Kongetira, P., Aingaran, K., Olukotun, K.: Niagara: A 32-way multithreaded Sparc processor. IEEE Micro, 25(2) (March/April 2005)
Zhirnov, V., Cavin, R.: Greg Leeming, Kosmas Galatsis, An Assessment of Integrated Digital Cellular Automata Architectures. IEEE Computer (January 2008)
Wu, W., Wen, M., Wu, N., He, Y., et al.: Research and Evaluating of a Multiple-dimension Scalable Stream Architecture. Acta Electronic Sinica (May 2008)
Ahn, J.H.: Memory and Control Organizaions of Stream Processors, Ph.D. Thesis, Stanford University (2007)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Wu, N. et al. (2011). Tiled Multi-Core Stream Architecture. In: Stenström, P. (eds) Transactions on High-Performance Embedded Architectures and Compilers IV. Lecture Notes in Computer Science, vol 6760. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-24568-8_14
Download citation
DOI: https://doi.org/10.1007/978-3-642-24568-8_14
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-24567-1
Online ISBN: 978-3-642-24568-8
eBook Packages: Computer ScienceComputer Science (R0)