AVR: Reducing Memory Traffic with Approximate Value Reconstruction
Paper in proceedings, 2019

This paper describes Approximate Value Reconstruction (AVR), an architecture for approximate memory compression. AVR reduces the memory traffic of applications that tolerate approximations in their dataset. Thereby, it utilizes more efficiently off-chip bandwidth improving significantly system performance and energy efficiency. AVR compresses memory blocks using low latency downsampling that exploits similarities between neighboring values and achieves aggressive compression ratios, up to 16:1 in our implementation. The proposed AVR architecture supports our compression scheme maximizing its effect and minimizing its overheads by (i) co-locating in the Last Level Cache (LLC) compressed and uncompressed data, (ii) efficiently handling LLC evictions, (iii) keeping track of badly compressed memory blocks, and (iv) avoiding LLC pollution with unwanted decompressed data. For applications that tolerate aggressive approximation in large fractions of their data, AVR reduces memory traffic by up to 70%, execution time by up to 55%, and energy costs by up to 20% introducing less than 1% error to the application output.

Compression

Approximate Computing

Cache Compression

Author

Albin Eldstål Damlin

Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers), Computer Systems

Pedro Petersen Moura Trancoso

Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers)

Ioannis Sourdis

Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers)

ACM International Conference Proceeding Series

Vol. 5 August 2019 a4

48th International Conference on Parallel Processing, ICPP 2019
Kyoto, Japan,

ACE: Approximate Algorithms and Computing Systems

Swedish Research Council (VR), 2015-01-01 -- 2018-12-31.

Subject Categories

Computer Engineering

Communication Systems

Computer Systems

Areas of Advance

Information and Communication Technology

Infrastructure

C3SE (Chalmers Centre for Computational Science and Engineering)

DOI

10.1145/3337821.3337824

More information

Latest update

4/28/2020