Architecting an Energy-Efficient DRAM System for GPUs
Authors:
- Niladrish ChatterjeeNCNiladrish Chatterjee
- Mike OconnorMOMike Oconnor
- Donghyuk LeeDLDonghyuk Lee
- Daniel R. JohnsonDJDaniel R. Johnson
- Stephen W. KecklerSKStephen W. Keckler
- Minsoo RhuMRMinsoo Rhu
- William J. DallyWDWilliam J. Dally
- Mike O’connorMOMike O’connor
- 1 February 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 73-84
Abstract
This paper proposes an energy-efficient, high-throughput DRAM architecture for GPUs and throughput processors. In these systems, requests from thousands of concurrent threads compete for a limited number of DRAM row buffers. As a result, only a fraction of the data fetched into a row buffer is used, leading to significant energy overheads. Our proposed DRAM architecture exploits the hierarchical organization of a DRAM bank to reduce the minimum row activation granularity. To avoid significant incremental area with this approach, we must partition the DRAM datapath into a number of semi-independent subchannels. These narrow subchannels increase data toggling energy which we mitigate using a static data reordering scheme designed to lower the toggle rate. This design has 35% lower energy consumption than a die-stacked DRAM with 2.6% area overhead. The resulting architecture, when augmented with an improved memory access protocol, can support parallel operations across the semi-independent subchannels, thereby improving system performance by 13% on average for a range of workloads.Keywords
This publication has 29 references indexed in Scilit:
- Page Placement Strategies for GPUs within Heterogeneous Memory Systems2015
- Half-DRAMACM SIGARCH Computer Architecture News, 2014
- A locality-aware memory hierarchy for energy-efficient GPU architectures2013
- System and Circuit Level Power Modeling of Energy-Efficient 3D-Stacked Wide I/O DRAMs2013
- Base-delta-immediate compression2012
- Staged memory schedulingACM SIGARCH Computer Architecture News, 2012
- Improving System Energy Efficiency with Memory Rank SubsettingACM Transactions on Architecture and Code Optimization, 2012
- Reducing memory interference in multicore systems via application-aware memory channel partitioning2011
- Future scaling of processor-memory interfaces2009
- A 1.6-GB/s data-rate 1-Gb synchronous DRAM with hierarchical square-shaped memory block and distributed bank architectureIEEE Journal of Solid-State Circuits, 1996