Reference distance as a metric for data locality
- Authors
- Pyo, C; Lee, G; Lee, KW; Han, HK
- Issue Date
- 1997
- Publisher
- I E E E, COMPUTER SOC PRESS
- Citation
- HIGH PERFORMANCE COMPUTING ON THE INFORMATION SUPERHIGHWAY - HPC ASIA '97, PROCEEDINGS, pp.151 - 156
- Journal Title
- HIGH PERFORMANCE COMPUTING ON THE INFORMATION SUPERHIGHWAY - HPC ASIA '97, PROCEEDINGS
- Start Page
- 151
- End Page
- 156
- URI
- https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/27665
- DOI
- 10.1109/HPC.1997.592139
- Abstract
- We propose reference distance as a metric for data locality. Reference distance is the number of referenced memory blocks between two successive references to the same memory block. Effectiveness of program transformations for data locality can be evaluated by using reference distance. Our claim is demonstrated by showing the change in data locality and speedup of matrix multiplication due to loop interchange, loop tiling, and loop unrolling. We also present a result of an experiment using loop distribution and unrolling with respect to two subroutines of Perfect benchmark programs. Reference distance would serve as a useful tool for developing program transformation scheme for data locality optimization.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/27665)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.