Building packet buffers using interleaved memories

High end routers need to store a large amount of data. Dynamic random access memories (DRAMs) are typically used for this purpose. However, DRAM memory devices don't match the bandwidth requirements, especially in terms of random access speeds. In this paper, we analyze a generalized memory interleaving scheme. This scheme implements a large, fast memory using multiple, slower DRAMs. In the presence of small amount of speed-up, we show that reasonable statistical guarantees (i.e., low drop probabilities) can be provided by using small SRAM buffers that queue read/write requests to DRAMs. We then relate drop probabilities to SRAM buffer size for a wide range of statistical arrival patterns.

[1]  Cheng Song,et al.  High performance TCP in ANSNET , 1994, CCRV.

[2]  A. Odlyzko,et al.  Internet growth: is there a Moore's law for data traffic? , 2000 .

[3]  William J. Dally,et al.  Memory access scheduling , 2000, Proceedings of 27th International Symposium on Computer Architecture (IEEE Cat. No.RS00201).

[4]  David A. Patterson,et al.  Computer Architecture - A Quantitative Approach (4. ed.) , 2007 .

[5]  Sally A. McKee,et al.  Access order and effective bandwidth for streams on a Direct Rambus memory , 1999, Proceedings Fifth International Symposium on High-Performance Computer Architecture.

[6]  David A. Patterson,et al.  Computer Architecture: A Quantitative Approach , 1969 .

[7]  Sally A. McKee,et al.  Design of a parallel vector access unit for SDRAM memory systems , 2000, Proceedings Sixth International Symposium on High-Performance Computer Architecture. HPCA-6 (Cat. No.PR00550).

[8]  Jin Cao,et al.  A Poisson limit for buffer overflow probabilities , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[9]  Nick McKeown,et al.  Doubling memory bandwidth for network buffers , 1998, Proceedings. IEEE INFOCOM '98, the Conference on Computer Communications. Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies. Gateway to the 21st Century (Cat. No.98.

[10]  Sally A. McKee,et al.  Access ordering and memory-conscious cache utilization , 1995, Proceedings of 1995 1st IEEE Symposium on High Performance Computer Architecture.

[12]  Mateo Valero,et al.  Command vector memory systems: high performance at low cost , 1998, Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192).

[13]  Wei-Fen Lin,et al.  Reducing DRAM latencies with an integrated memory hierarchy design , 2001, Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture.

[14]  Guido Appenzeller,et al.  Sizing router buffers , 2004, SIGCOMM '04.

[15]  Thomas Alexander,et al.  Distributed prefetch-buffer/cache design for high performance memory systems , 1996, Proceedings. Second International Symposium on High-Performance Computer Architecture.