Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/41069
Title: Adaptive compiler directed prefetching for EPIC processors
Authors: Kim, J.
Rabbah, R.M.
Palem, K.V.
Wong, W.-F. 
Keywords: Compiler Directed Data Prefetching
EPIC Architecture
Markovian Predictor
Memory Bottleneck
Off-line Learning
Issue Date: 2004
Source: Kim, J.,Rabbah, R.M.,Palem, K.V.,Wong, W.-F. (2004). Adaptive compiler directed prefetching for EPIC processors. Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04 1 : 495-501. ScholarBank@NUS Repository.
Abstract: The widely acknowledged performance gap between processors and memory has been the subject of much research. In the Explicitly Parallel Instruction Computing (EPIC) paradigm, the combination of in-order issue and the presence of a large number of parallel functional units exacerbate the problem. Prefetching, by hardware, software, or a combination of both, is one of the primary mechanisms advocated to alleviate this problem. In this paper, we propose a new mechanism readily suitable for implementation in EPIC processors. Specifically, we introduce a predicated prefetch operation which leverages the concept of an informing load to dynamically adapt to run-time memory behaviors. Furthermore, we employ predicated Prefetching in a new optimization framework, which also consists of data remapping and off-line learning of Markovian predictors. This distinguishes our approach from early software Prefetching techniques that only involve static program analysis. Our experiments show that the proposed framework can effectively remove 10%-30% of the stall cycles due to cache misses for benchmarks from the well-known SPEC and OLDEN suites.
Source Title: Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA'04
URI: http://scholarbank.nus.edu.sg/handle/10635/41069
ISBN: 1932415262
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

54
checked on Dec 9, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.