Options
QSLC: Quantization-Based, Low-Error Selective Approximation for GPUs
Publikationstyp
Conference Paper
Date Issued
2021-02
Sprache
English
Author(s)
Volume
2021-February
Start Page
475
End Page
480
Article Number
9474124
Citation
Design, Automation and Test in Europe Conference and Exhibition (DATE 2021)
Contribution to Conference
Publisher DOI
Scopus ID
GPUs use a large memory access granularity (MAG) that often results in a low effective compression ratio for memory compression techniques. The low effective compression ratio is caused by a significant fraction of compressed blocks that have a few bytes above a multiple of MAG. While MAG-aware selective approximation, based on a tree structure, has been used to increase the effective compression ratio and the performance gain, approximation results in a high error that is reduced by using complex optimizations. We propose a simple quantization-based approximation technique (QSLC) that can also selectively approximate a few bytes above MAG. While the quantization-based approximation technique has a similar performance to the state-of-the-art tree-based selective approximation, the average error for the quantization-based technique is 5× lower. We further trade-off the two techniques and show that the area and power overhead of the quantization-based technique is 12.1× and 7.6× lower than the state-of-the-art, respectively. Our sensitivity analysis to different block sizes further shows the opportunities and the significance of MAG-aware selective approximation.