EUVIMEDEUVIMED entdecken
Crossref

RIMMS: Runtime Integrated Memory Management System for Heterogeneous Computing

Serhan Gener, Aditya Ukarande, Shilpa Mysore Srinivasa murthy, Sahil Hassan, Joshua Mack, Chaitali Chakrabarti, Umit Ogras, Ali Akoglu

ACM Transactions on Embedded Computing Systems · 30.11.2025 · 24 (5s) · 1-24

DOI: 10.1145/3760257 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Embedded Computing Systems, online erschienen am 26.09.2025. Verfasst von Serhan Gener und 7 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: Air Force Research Laboratory (AFRL) and Defense Advanced Research Projects Agency.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Efficient memory management in heterogeneous systems is increasingly challenging due to diverse compute architectures (e.g., CPU, GPU, and FPGA) and dynamic task mappings not known at compile time. Existing approaches often require programmers to manage data placement and transfers explicitly, or assume static mappings that limit portability and scalability. This article introduces RIMMS (Runtime Integrated Memory Management System), a lightweight, runtime-managed, hardware-agnostic memory abstraction layer that decouples application development from low-level memory operations. RIMMS transparently tracks data locations, manages consistency, and supports efficient memory allocation across heterogeneous compute elements without requiring platform-specific tuning or code modifications. We integrate RIMMS into a baseline runtime and evaluate with complete radar signal processing applications across CPU+GPU and CPU+FPGA platforms. RIMMS delivers up to 2.43× speedup on GPU-based and 1.82× on FPGA-based systems over the baseline. Compared to IRIS, a recent heterogeneous runtime system, RIMMS achieves up to 3.08X speedup and matches the performance of native CUDA implementations while significantly reducing programming complexity. Despite operating at a higher abstraction level, RIMMS incurs only 1–2 cycles of overhead per memory management call, making it a low-cost solution. These results demonstrate RIMMS’s ability to deliver high performance and enhanced programmer productivity in dynamic, real-world heterogeneous environments.

Förderung

  • Air Force Research Laboratory (AFRL) and Defense Advanced Research Projects Agency

Zitieren

APA

Gener, S., Ukarande, A., murthy, S. M. S., Hassan, S., Mack, J., Chakrabarti, C., Ogras, U., & Akoglu, A. (2025). RIMMS: Runtime Integrated Memory Management System for Heterogeneous Computing. ACM Transactions on Embedded Computing Systems, 24(5s), 1-24. https://doi.org/10.1145/3760257

Vancouver

Gener S, Ukarande A, murthy SMS, Hassan S, Mack J, Chakrabarti C, et al. RIMMS: Runtime Integrated Memory Management System for Heterogeneous Computing. ACM Transactions on Embedded Computing Systems. 2025;24(5s):1-24. doi:10.1145/3760257

BibTeX
@article{Gener2025,
  author = {Gener, Serhan and Ukarande, Aditya and murthy, Shilpa Mysore Srinivasa and Hassan, Sahil and Mack, Joshua and Chakrabarti, Chaitali and Ogras, Umit and Akoglu, Ali},
  title = {RIMMS: Runtime Integrated Memory Management System for Heterogeneous Computing},
  journal = {ACM Transactions on Embedded Computing Systems},
  year = {2025},
  volume = {24},
  number = {5s},
  pages = {1--24},
  doi = {10.1145/3760257}
}
Kostenlos starten →