Application Specific Memory Access, Reuse and Reordering for SDRAM

Abstract

AbstractThe efficient use of bandwidth available on an external SDRAM interface is strongly dependent on the sequence of addresses requested. On-chip memory buffers can make possible data reuse and request reordering which together ensure bandwidth on an SDRAM interface is used efficiently. This paper outlines an automated procedure for generating an application-specific memory hierarchy which exploits reuse and reordering and quantifies the impact this has on memory bandwidth over a range of representative benchmarks. Considering a range of parameterized designs, we observe up to 50x reduction in the quantity of data fetched from external memory. This, combined with reordering of the transactions, allows up to 128x reduction in the memory access time of certain memory-intensive benchmarks.KeywordsMemory AccessLoop NestExternal MemoryMemory BandwidthMemory AddressThese 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.