Proceedings 2000 IEEE Symposium on Field-Programmable Custom Computing Machines (Cat. No.PR00871)
DOI: 10.1109/fpga.2000.903390
|View full text |Cite
|
Sign up to set email alerts
|

Configuration caching management techniques for reconfigurable computing

Abstract: Although run-time reconfigurable systems have been shown to achieve very high performance, the speedups over traditional microprocessor systems are limited by the cost of configuration of the hardware. In this paper, we explore the idea of configuration caching, and create some of the first cache management algorithms for reconfigurable systems. We present techniques to carefully manage the configurations present on the reconfigurable hardware throughout program execution. Through the use of the presented stra… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

1
88
0

Publication Types

Select...
5
3
2

Relationship

0
10

Authors

Journals

citations
Cited by 93 publications
(93 citation statements)
references
References 13 publications
1
88
0
Order By: Relevance
“…To minimize delay, the goal would be to group together states which typically execute together, so as to minimize the frequency of state transitions that cross the CP boundary. In their work on configuration caching, Li et al [48] developed a clustering operation which attacks a problem very close to the area-and IO-bound state packing problem that would be necessary for the general CP mapping case.…”
Section: Page Mappingmentioning
confidence: 99%
“…To minimize delay, the goal would be to group together states which typically execute together, so as to minimize the frequency of state transitions that cross the CP boundary. In their work on configuration caching, Li et al [48] developed a clustering operation which attacks a problem very close to the area-and IO-bound state packing problem that would be necessary for the general CP mapping case.…”
Section: Page Mappingmentioning
confidence: 99%
“…Their hardware reconfigurability allows limited resources to perform multiple functions, and satisfies both flexibility and performance. Although reconfiguration overhead in most RAs is reported serious [1]- [3], configuring a kernel and executing it thousands of times could amortize the overhead. However, in RAs that have relatively small number of resources just like coarse-grained RAs, some loops cannot be implemented at one time due to lack of available resources and need to be divided and to be configured repeatedly [4].…”
Section: Introductionmentioning
confidence: 99%
“…These studies include configuration compression [6] and configuration caching [7]. Another promising technique that has been studied is configuration prefetching [5] [4].…”
Section: Introductionmentioning
confidence: 99%