1989 IEEE International Conference on Computer-Aided Design. Digest of Technical Papers
DOI: 10.1109/iccad.1989.76897
|View full text |Cite
|
Sign up to set email alerts
|

Scheduling and hardware sharing in pipelined data paths

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
19
0
2

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 72 publications
(21 citation statements)
references
References 7 publications
0
19
0
2
Order By: Relevance
“…Pipelining is an effective method to optimize the execution of a loop with or without loop carried dependencies, especially for DSP [8]. Highly concurrent implementations can be obtained by overlapping the execution of consecutive iterations.…”
Section: Pipeline Synthesis and Optimization: Backgroundmentioning
confidence: 99%
See 1 more Smart Citation
“…Pipelining is an effective method to optimize the execution of a loop with or without loop carried dependencies, especially for DSP [8]. Highly concurrent implementations can be obtained by overlapping the execution of consecutive iterations.…”
Section: Pipeline Synthesis and Optimization: Backgroundmentioning
confidence: 99%
“…Sehwa minimizes time delay using a modified list scheduling algorithm with a resource allocation table. HAL [7] performs a time-constrained, functional pipelining scheduling using the force directed method which is modified in [8]. The loop winding method was proposed in the Elf [9] system.…”
Section: Pipeline Synthesis and Optimization: Backgroundmentioning
confidence: 99%
“…A graph partitioning technique for scheduling a pipelined data path was presented in [23]. The force-directed scheduling algorithm has also been adapted to solve the same problem in [3], [21].…”
Section: Considerations In Data-path Synthesismentioning
confidence: 99%
“…The operations not on the critical path are assigned one at a time according to their degree of freedom. In force-directed scheduling [3], [21], "force" values are calculated for all operations at all feasible control steps. The pairing of operation and control step that has the most attractive force is selected and assigned.…”
Section: Basic Scheduling Techniquesmentioning
confidence: 99%
“…Many efforts in high-level scheduling have been concentrated on improving circuit performance (time required to execute all the behavioural description operations) by minimizing the slack times wasted in clock cycles. Traditionally, pipelining has been the preferred technique to improve system performance, although it does not reduce the circuit latency [1][2]. In order to reduce the latency, some algorithms have added some optimization phases after the scheduling process to adjust either the number or duration of the clock cycles [3][4][5][6].…”
Section: Introductionmentioning
confidence: 99%