Selective DRAM cache bypassing for improving bandwidth on DRAM/NVM hybrid main memory systems

DOI Web Site 2 Citations 4 References
  • Ro Yuhwan
    Department of Transdisciplinary Studies, Seoul National University
  • Sung Minchul
    Department of Transdisciplinary Studies, Seoul National University
  • Park Yongjun
    Department of Computer Science and Engineering, Hanyang University
  • Ahn Jung Ho
    Department of Transdisciplinary Studies, Seoul National University

Search this article

Description

<p>Satisfying a demand for higher memory capacity is a major problem for computing systems. Conventional solutions are reaching those limits; instead, DRAM/NVM hybrid main memory systems which consist of emerging Non-Volatile Memory for large capacity and DRAM last-level cache for high access speed were proposed for further improvement. However, in these systems, the two device types share limited memory channels/ranks and NVM channels/ranks are often less utilized than DRAM ones. This paper proposes an OBYST (On hit BYpass to STeal bandwidth) technique to improve memory bandwidth by selectively sending read requests that hit on DRAM cache to NVM instead of busy DRAM. We also propose an inter-device request scheduling policy optimized to OBYST. With negligible area overhead, OBYST improves bandwidth, IPC, and EDP by up to 22%, 21%, and 26% over the baseline without bandwidth optimizations, respectively.</p>

Journal

  • IEICE Electronics Express

    IEICE Electronics Express 14 (11), 20170437-20170437, 2017

    The Institute of Electronics, Information and Communication Engineers

Citations (2)*help

See more

References(4)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top