2010 IEEE International Conference on Computer Design 2010
DOI: 10.1109/iccd.2010.5647740
|View full text |Cite
|
Sign up to set email alerts
|

Predicting the throughput of multiprocessor applications under dynamic workload

Abstract: Abstract-This work contributes to throughput calculation for real-time multiprocessor applications experiencing dynamic workload variations. We focus on a method to predict the system throughput when processing an arbitrarily long data frame given the meta-characteristics of the workload in that frame. This is useful for different purposes, such as resource allocation or dynamic voltage scaling in embedded systems. An accurate enough analysis is not trivial when two factors are combined: parallelism and dynami… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2010
2010
2015
2015

Publication Types

Select...
5

Relationship

2
3

Authors

Journals

citations
Cited by 5 publications
(3 citation statements)
references
References 16 publications
0
3
0
Order By: Relevance
“…Various propositions can be found in [4], [5], [8], [12], [23], [26] or [27]. These and all static optimization decisions found in the literature are complementary to our runtime proposals.…”
Section: Related Workmentioning
confidence: 76%
“…Various propositions can be found in [4], [5], [8], [12], [23], [26] or [27]. These and all static optimization decisions found in the literature are complementary to our runtime proposals.…”
Section: Related Workmentioning
confidence: 76%
“…However, by limiting the vectors to pre-defined values, the size of the space-space can be limited. Heuristics for determining the invariant state vectors, can for instance allow only a single, universal vector, or may allow the state-vector to vary with the source and destination scenarios [28].…”
Section: A Throughput Analysismentioning
confidence: 99%
“…In a streaming, distributed embedded software system, the difficulty lies in the fact that different scenarios are concurrently active in different stages of a pipelined implementation. The techniques introduced in [9,21] and in this paper allow those scenarios to still be handled individually, compositionally. In this paper we introduce a method to derive the tightest possible performance guarantees for the class of SADF graphs.…”
Section: Introductionmentioning
confidence: 99%