2009 Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition 2009
DOI: 10.1109/date.2009.5090713
|View full text |Cite
|
Sign up to set email alerts
|

Fault-tolerant average execution time optimization for general-purpose multi-processor system-on-chips

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
14
0

Year Published

2010
2010
2014
2014

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 10 publications
(14 citation statements)
references
References 8 publications
0
14
0
Order By: Relevance
“…Checkpoint size and checkpoint interval was set in between 100-1000MB and 100-1000ms respectively. The results were obtained using the average of 5-10 experiments and than compared with the existing method RRC [2]. The Table1 reports the performance of CCRR scheme with assigned P, t and s. The results in Table1 indicate that CCRR scheme performs better than the RRC [2].…”
Section: Resultsmentioning
confidence: 94%
See 2 more Smart Citations
“…Checkpoint size and checkpoint interval was set in between 100-1000MB and 100-1000ms respectively. The results were obtained using the average of 5-10 experiments and than compared with the existing method RRC [2]. The Table1 reports the performance of CCRR scheme with assigned P, t and s. The results in Table1 indicate that CCRR scheme performs better than the RRC [2].…”
Section: Resultsmentioning
confidence: 94%
“…This time based CCRR uses a simple initialization procedure to start the checkpoint timers. Contrary to previous time based checkpointing [2], the new scheme also eliminates the overheads of intransit message storage and addition of information to message. This is accomplished by preventing processes from sending messages during an interval before the checkpoint time and the CCRR technique leads to faster execution time with better performance.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…RRC has been the subject of research for both non-real-time [1], [2] [3], [4] and real-time systems [5], [6], [7], [8], [9]. While for non-real-time systems, it is important to minimize the average execution time when RRC is applied, it is for real-time systems important to maximize the probability that a job meets a given deadline, [10].…”
Section: Amentioning
confidence: 99%
“…The results presented in Table IV and Table V are acquired by running all combinations of values for n c and k, while the results from the optimization method only require two iterations for Scenario A, and eight iterations for Scenario B (observe the number of re-executions, k, in Table IV and Table V for n c = 10 and n c = 20 respectively). For P3, we consider an RRC optimization approach for a non-real-time system that provides an optimal number of checkpoints, n * c , that leads to minimal average execution time (AET ), [4]. For Scenario A, the approach, [4], computes n * c = 1 and a minimal AET = 1020t.u., while for Scenario B, the approach computes n * c = 3 and a minimal AET = 1138t.u.…”
Section: Rmentioning
confidence: 99%