BEAR
- 13 June 2015
- conference paper
- conference paper
- Published by Association for Computing Machinery (ACM) in Proceedings of the 42nd Annual International Symposium on Computer Architecture
- p. 198-210
- https://doi.org/10.1145/2749469.2750387
Abstract
Die stacking memory technology can enable gigascale DRAM caches that can operate at 4x-8x higher bandwidth than commodity DRAM. Such caches can improve system performance by servicing data at a faster rate when the requested data is found in the cache, potentially increasing the memory bandwidth of the system by 4x-8x. Unfortunately, a DRAM cache uses the available memory bandwidth not only for data transfer on cache hits, but also for other secondary operations such as cache miss detection, fill on cache miss, and writeback lookup and content update on dirty evictions from the last-level on-chip cache. Ideally, we want the bandwidth consumed for such secondary operations to be negligible, and have almost all the bandwidth be available for transfer of useful data from the DRAM cache to the processor. We evaluate a 1GB DRAM cache, architected as Alloy Cache, and show that even the most bandwidth-efficient proposal for DRAM cache consumes 3.8x bandwidth compared to an idealized DRAM cache that does not consume any bandwidth for secondary operations. We also show that redesigning the DRAM cache to minimize the bandwidth consumed by secondary operations can potentially improve system performance by 22%. To that end, this paper proposes Bandwidth Efficient ARchitecture (BEAR) for DRAM caches. BEAR integrates three components, one each for reducing the bandwidth consumed by miss detection, miss fill, and writeback probes. BEAR reduces the bandwidth consumption of DRAM cache by 32%, which reduces cache hit latency by 24% and increases overall system performance by 10%. BEAR, with negligible overhead, outperforms an idealized SRAM Tag-Store design that incurs an unacceptable overhead of 64 megabytes, as well as Sector Cache designs that incur an SRAM storage overhead of 6 megabytes.Keywords
Funding Information
- Defense Advanced Research Projects Agency
- Semiconductor Research Corporation
This publication has 20 references indexed in Scilit:
- ATCachePublished by Association for Computing Machinery (ACM) ,2014
- Insertion and promotion for tree-based PseudoLRU last-level cachesPublished by Association for Computing Machinery (ACM) ,2013
- Die-stacked DRAM caches for serversPublished by Association for Computing Machinery (ACM) ,2013
- Fundamental Latency Trade-off in Architecting DRAM Caches: Outperforming Impractical SRAM-Tags with a Simple and Practical DesignPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2012
- A Mostly-Clean DRAM Cache for Effective Hit Speculation and Self-Balancing DispatchPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2012
- Efficiently enabling conventional block sizes for very large die-stacked DRAM cachesPublished by Association for Computing Machinery (ACM) ,2011
- Design and optimization of large size and low overhead off-chip cachesInternational Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2004
- Using SimPoint for accurate and efficient simulationPublished by Association for Computing Machinery (ACM) ,2003
- Dead-block prediction & dead-block correlating prefetchersPublished by Association for Computing Machinery (ACM) ,2001
- A modified approach to data cache managementPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1995