GBDI: Going Beyond Base-Delta-Immediate Compression with Global Bases
Paper i proceeding, 2022

Memory bandwidth is limiting performance for many emerging applications. While compression techniques can unlock a higher memory bandwidth, prior art offers only modestly better bandwidth. This paper contributes with a new compression method - Global Base Delta Immediate compression (GBDI) - that offers substantially higher memory bandwidth by, unlike prior art, selecting base values across memory blocks. GBDI uses a novel clustering algorithm through data analysis in the background. The presented accelerator infrastructure offers low area overhead and latency. This paper shows that GBDI offers a compression ratio of 2.3×, and yields 1.5× higher bandwidth and 1.1× higher performance compared with a baseline without compression support, on average, for SPEC2017 benchmarks requiring medium to high memory bandwidth.

Författare

Alexandra Angerd

ZeroPoint Technologies

Angelos Arelakis

ZeroPoint Technologies

Vasilis Spiliopoulos

ZeroPoint Technologies

Erik Sintorn

Embedded Electronics Systems and Computer Graphics

Per Stenström

Chalmers, Data- och informationsteknik, Dator- och nätverkssystem

ZeroPoint Technologies

Proceedings - International Symposium on High-Performance Computer Architecture

15300897 (ISSN)

Vol. 2022-April 1115-1127
9781665420273 (ISBN)

28th Annual IEEE International Symposium on High-Performance Computer Architecture, HPCA 2022
Virtual, Online, South Korea,

Principer för beräknande minnesenheter (PRIDE)

Stiftelsen för Strategisk forskning (SSF) (DnrCHI19-0048), 2021-01-01 -- 2025-12-31.

Ämneskategorier

Datorteknik

Kommunikationssystem

Datavetenskap (datalogi)

DOI

10.1109/HPCA53966.2022.00085

ISBN

9781665420273

Mer information

Senast uppdaterat

2022-08-24