EUVIMEDEUVIMED entdecken
Crossref

Enhancing the Performance of Learned Index on Processing-in-Memory Architecture

Lixiao Cui, Kedi Yang, Yusen Li, Gang Wang, Xiaoguang Liu

ACM Transactions on Storage · 15.07.2026

DOI: 10.1145/3829362 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Storage, online erschienen am 15.07.2026. Verfasst von Lixiao Cui und 4 weiteren Autor:innen. Die Arbeit nennt 72 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Learned indexes serve as essential index structures within storage systems, yet their efficiency is constrained by the memory bandwidth and latency limitations inherent in traditional computer architectures. Processing-in-memory (PIM) technology presents a promising approach by embedding processing units directly into memory components. In this paper, we introduce PIMLex, a well-designed learned index utilizing PIM to address the challenges posed by memory-bound operations. PIMLex tackles the capacity constraints found in current PIM hardware by adopting a decoupled two-level architecture, which harnesses both the robust data processing power of PIM and the expansive capacity of conventional DRAM. Furthermore, the design incorporates a PIM-optimized model structure to reduce computational demands that are inefficient for PIM. Coupled with a hotness-aware replication strategy that promotes load balancing across multiple PIM modules, PIMLex achieves superior performance under diverse workload scenarios. We have implemented PIMLex on UPMEM, a commercially available PIM platform, demonstrating a throughput improvement of 2.2 × compared to the DRAM-based ALEX.

Zitieren

APA

Cui, L., Yang, K., Li, Y., Wang, G., & Liu, X. (2026). Enhancing the Performance of Learned Index on Processing-in-Memory Architecture. ACM Transactions on Storage. https://doi.org/10.1145/3829362

Vancouver

Cui L, Yang K, Li Y, Wang G, Liu X. Enhancing the Performance of Learned Index on Processing-in-Memory Architecture. ACM Transactions on Storage. 2026. doi:10.1145/3829362

BibTeX
@article{Cui2026,
  author = {Cui, Lixiao and Yang, Kedi and Li, Yusen and Wang, Gang and Liu, Xiaoguang},
  title = {Enhancing the Performance of Learned Index on Processing-in-Memory Architecture},
  journal = {ACM Transactions on Storage},
  year = {2026},
  doi = {10.1145/3829362}
}
Kostenlos starten →