IMP: Indirect memory prefetcher

Machine learning, graph analytics and sparse linear algebra-based applications are dominated by irregular memory accesses resulting from following edges in a graph or non-zero elements in a sparse matrix. These accesses have little temporal or spatial locality, and thus incur long memory stalls and...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2015 48th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO) S. 178 - 190
Hauptverfasser: Xiangyao Yu, Hughes, Christopher J., Satish, Nadathur, Devadas, Srinivas
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: ACM 01.12.2015
Schlagworte:
Arrays
Bandwidth
Hardware
Indexes
Multicore processing
Prefetching
Sparse matrices
ISSN:2379-3155
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Machine learning, graph analytics and sparse linear algebra-based applications are dominated by irregular memory accesses resulting from following edges in a graph or non-zero elements in a sparse matrix. These accesses have little temporal or spatial locality, and thus incur long memory stalls and large bandwidth requirements. A traditional streaming or striding prefetcher cannot capture these irregular access patterns. A majority of these irregular accesses come from indirect patterns of the form A[B[j]]. We propose an efficient hardware indirect memory prefetcher (IMP) to capture this access pattern and hide latency. We also propose a partial cacheline accessing mechanism for these prefetches to reduce the network and DRAM bandwidth pressure from the lack of spatial locality. Evaluated on 7 applications, IMP shows 56% speedup on average (up to 2.3×) compared to a baseline 64 core system with streaming prefetchers. This is within 23% of an idealized system. With partial cacheline accessing, we see another 9.4% speedup on average (up to 46.6%).
ISSN:2379-3155
DOI:10.1145/2830772.2830807