Analysis of Data Transfer Bottlenecks in Commercial PIM Systems: A Study With UPMEM-PIM

Analysis of Data Transfer Bottlenecks in Commercial PIM Systems: A Study With UPMEM-PIM

Due to emerging workloads that require high memory bandwidth, Processing-in-Memory (PIM) has gained significant attention and led several industrial PIM products to be introduced which are integrated with conventional computing systems. This letter characterizes the data transfer overheads between c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE computer architecture letters 2024-07, Vol.23 (2), p.179-182
Hauptverfasser: Lee, Dongjae, Hyun, Bongjoon, Kim, Taehun, Rhu, Minsoo
Format: Artikel
Sprache:eng
Schlagworte:
Bandwidth
Data transfer
Memory management
near-memory processing
parallel architecture
Processing-in-memory (PIM)
Random access memory
Runtime library
Software
Throughput
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Due to emerging workloads that require high memory bandwidth, Processing-in-Memory (PIM) has gained significant attention and led several industrial PIM products to be introduced which are integrated with conventional computing systems. This letter characterizes the data transfer overheads between conventional DRAM address space and PIM address space within a PIM-integrated system using the commercialized PIM device made by UPMEM. Our findings highlight the need for optimization in PIM-integrated systems to address these overheads, offering critical insights for future PIM technologies.
ISSN:1556-6056
1556-6064
DOI:10.1109/LCA.2024.3387472