EUVIMEDEUVIMED entdecken
Crossref

Asynchronous Memory Access Unit: Exploiting Massive Parallelism for Far Memory Access

Luming Wang, Xu Zhang, Songyue Wang, Zhuolun Jiang, Tianyue Lu, Mingyu Chen, Siwei Luo, Keji Huang

ACM Transactions on Architecture and Code Optimization · 30.09.2024 · 21 (3) · 1-28

DOI: 10.1145/3663479 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 14.09.2024. Verfasst von Luming Wang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 74 Literaturangaben. Gefördert von: National Key Research and Development Program of China NKRDPC.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The growing memory demands of modern applications have driven the adoption of far memory technologies in data centers to provide cost-effective, high-capacity memory solutions. However, far memory presents new performance challenges because its access latencies are significantly longer and more variable than local DRAM. For applications to achieve acceptable performance on far memory, a high degree of memory-level parallelism (MLP) is needed to tolerate the long access latency. While modern out-of-order processors are capable of exploiting a certain degree of MLP, they are constrained by resource limitations and hardware complexity. The key obstacle is the synchronous memory access semantics of traditional load/store instructions, which occupy critical hardware resources for a long time. The longer far memory latencies exacerbate this limitation. This article proposes a set of Asynchronous Memory Access Instructions (AMI) and its supporting function unit, Asynchronous Memory Access Unit (AMU), inside contemporary Out-of-Order Core. AMI separates memory request issuing from response handling to reduce resource occupation. Additionally, AMU architecture supports up to several hundreds of asynchronous memory requests through re-purposing a portion of L2 Cache as scratchpad memory (SPM) to provide sufficient temporal storage. Together with a coroutine-based programming framework, this scheme can achieve significantly higher MLP for hiding far memory latencies. Evaluation with a cycle-accurate simulation shows AMI achieves 2.42× speedup on average for memory-bound benchmarks with 1μs additional far memory latency. Over 130 outstanding requests are supported with 26.86× speedup for GUPS (random access) with 5 μs latency. These demonstrate how the techniques tackle far memory performance impacts through explicit MLP expression and latency adaptation.

Förderung

  • National Key Research and Development Program of China NKRDPC

Zitieren

APA

Wang, L., Zhang, X., Wang, S., Jiang, Z., Lu, T., Chen, M., Luo, S., & Huang, K. (2024). Asynchronous Memory Access Unit: Exploiting Massive Parallelism for Far Memory Access. ACM Transactions on Architecture and Code Optimization, 21(3), 1-28. https://doi.org/10.1145/3663479

Vancouver

Wang L, Zhang X, Wang S, Jiang Z, Lu T, Chen M, et al. Asynchronous Memory Access Unit: Exploiting Massive Parallelism for Far Memory Access. ACM Transactions on Architecture and Code Optimization. 2024;21(3):1-28. doi:10.1145/3663479

BibTeX
@article{Wang2024,
  author = {Wang, Luming and Zhang, Xu and Wang, Songyue and Jiang, Zhuolun and Lu, Tianyue and Chen, Mingyu and Luo, Siwei and Huang, Keji},
  title = {Asynchronous Memory Access Unit: Exploiting Massive Parallelism for Far Memory Access},
  journal = {ACM Transactions on Architecture and Code Optimization},
  year = {2024},
  volume = {21},
  number = {3},
  pages = {1--28},
  doi = {10.1145/3663479}
}
Kostenlos starten →