Traditional graphics processing units (GPUs) suffer from the low memory capacity and demand for high memory bandwidth. To address these challenges, we propose Ohm-GPU, a new optical network based heterogeneous memory design for GPUs. Specifically, Ohm-GPU can expand the memory capacity by combing a set of high-density 3D XPoint and DRAM modules as heterogeneous memory. To prevent memory channels from throttling throughput of GPU memory system, Ohm-GPU replaces the electrical lanes in the traditional memory channel with a high-performance optical network. However, the hybrid memory can introduce frequent data migrations betweenDRAMand 3D XPoint, which can unfortunately occupy the memory channel and increase the optical network traffic. To prevent the intensive data migrations from blocking normal memory services, Ohm-GPU revises the existing memory controller and designs a new optical network infrastructure, which enables the memory channel to serve the data migrations and memory requests, in parallel. Our evaluation results reveal that Ohm-GPU can improve the performance by 181% and 27%, compared to a DRAMbased GPU memory system and the baseline optical network based heterogeneous memory system, respectively.
|Title of host publication||MICRO 2021 - 54th Annual IEEE/ACM International Symposium on Microarchitecture, Proceedings|
|Publisher||IEEE Computer Society|
|Number of pages||14|
|Publication status||Published - 2021 Oct 18|
|Event||54th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2021 - Virtual, Online, Greece|
Duration: 2021 Oct 18 → 2021 Oct 22
|Name||Proceedings of the Annual International Symposium on Microarchitecture, MICRO|
|Conference||54th Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2021|
|Period||21/10/18 → 21/10/22|
Bibliographical noteFunding Information:
This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by KAIST start-up package (G01190015), and Samsung (G01200447). Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Myoungsoo Jung is the corresponding author.
© 2021 Association for Computing Machinery.
All Science Journal Classification (ASJC) codes
- Hardware and Architecture