EUVIMEDEUVIMED entdecken
Crossref

MaPHeA: A Framework for Lightweight Memory Hierarchy-aware Profile-guided Heap Allocation

Deok-Jae Oh, Yaebin Moon, Do Kyu Ham, Tae Jun Ham, Yongjun Park, Jae W. Lee, Jung Ho Ahn, Eojin Lee

ACM Transactions on Embedded Computing Systems · 31.01.2023 · 22 (1) · 1-28

DOI: 10.1145/3527853 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Embedded Computing Systems, online erschienen am 13.12.2022. Verfasst von Deok-Jae Oh und 7 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 91 Literaturangaben. Gefördert von: R&D program of MOTIE/KEIT, Engineering Research Center Program through the National Research Foundation of Korea, Korean Government MSIT.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hardware performance monitoring units (PMUs) are a standard feature in modern microprocessors, providing a rich set of microarchitectural event samplers. Recently, numerous profile-guided optimization (PGO) frameworks have exploited them to feature much lower profiling overhead compared to conventional instrumentation-based frameworks. However, existing PGO frameworks mainly focus on optimizing the layout of binaries; they overlook rich information provided by the PMU about data access behaviors over the memory hierarchy. Thus, we propose MaPHeA, a lightweight M emory hierarchy- a ware P rofile-guided He ap A llocation framework applicable to both HPC and embedded systems. MaPHeA guides and applies the optimized allocation of dynamically allocated heap objects with very low profiling overhead and without additional user intervention to improve application performance. To demonstrate the effectiveness of MaPHeA, we apply it to optimizing heap object allocation in an emerging DRAM-NVM heterogeneous memory system (HMS), selective huge-page utilization, and controlling the cacheability of the objects with the low temporal locality. In an HMS, by identifying and placing frequently accessed heap objects to the fast DRAM region, MaPHeA improves the performance of memory-intensive graph-processing and Redis workloads by 56.0% on average over the default configuration that uses DRAM as a hardware-managed cache of slow NVM. By identifying large heap objects that cause frequent TLB misses and allocating them to huge pages, MaPHeA increases the performance of the read and update operations of Redis by 10.6% over the transparent huge-page implementation of Linux. Also, by distinguishing the objects that cause cache pollution due to their low temporal locality and applying write-combining to them, MaPHeA improves the performance of STREAM and RADIX workloads by 20.0% on average over the system without cacheability control.

Förderung

  • R&D program of MOTIE/KEIT
  • Engineering Research Center Program through the National Research Foundation of Korea
  • Korean Government MSIT
  • Inha University Research

Zitieren

APA

Oh, D., Moon, Y., Ham, D. K., Ham, T. J., Park, Y., Lee, J. W., Ahn, J. H., & Lee, E. (2023). MaPHeA: A Framework for Lightweight Memory Hierarchy-aware Profile-guided Heap Allocation. ACM Transactions on Embedded Computing Systems, 22(1), 1-28. https://doi.org/10.1145/3527853

Vancouver

Oh D, Moon Y, Ham DK, Ham TJ, Park Y, Lee JW, et al. MaPHeA: A Framework for Lightweight Memory Hierarchy-aware Profile-guided Heap Allocation. ACM Transactions on Embedded Computing Systems. 2023;22(1):1-28. doi:10.1145/3527853

BibTeX
@article{Oh2023,
  author = {Oh, Deok-Jae and Moon, Yaebin and Ham, Do Kyu and Ham, Tae Jun and Park, Yongjun and Lee, Jae W. and Ahn, Jung Ho and Lee, Eojin},
  title = {MaPHeA: A Framework for Lightweight Memory Hierarchy-aware Profile-guided Heap Allocation},
  journal = {ACM Transactions on Embedded Computing Systems},
  year = {2023},
  volume = {22},
  number = {1},
  pages = {1--28},
  doi = {10.1145/3527853}
}
Kostenlos starten →