2008 IEEE Fourth International Conference on eScience 2008
DOI: 10.1109/escience.2008.66
|View full text |Cite
|
Sign up to set email alerts
|

Resource Availability Prediction for Improved Grid Scheduling

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

0
15
0

Year Published

2010
2010
2012
2012

Publication Types

Select...
4
1
1

Relationship

1
5

Authors

Journals

citations
Cited by 28 publications
(15 citation statements)
references
References 11 publications
0
15
0
Order By: Relevance
“…Importantly and as described in our previous work, the differences in our prediction techniques lead to our predictors producing improved prediction accuracies. Our prediction techniques analyze different portions of a resource's history to enable predictions and calculate those predictions with an improved technique (including a transition weighting scheme) which has been shown to produce higher prediction accuracies [6]. Previous papers [3] [7] have described in detail the related work in the areas of availability analysis (most notably Ren et.…”
Section: Related Workmentioning
confidence: 99%
See 2 more Smart Citations
“…Importantly and as described in our previous work, the differences in our prediction techniques lead to our predictors producing improved prediction accuracies. Our prediction techniques analyze different portions of a resource's history to enable predictions and calculate those predictions with an improved technique (including a transition weighting scheme) which has been shown to produce higher prediction accuracies [6]. Previous papers [3] [7] have described in detail the related work in the areas of availability analysis (most notably Ren et.…”
Section: Related Workmentioning
confidence: 99%
“…Moreover, resource availability predictors can use past (and potentially group) behavior to help forecast whether a resource will become unavailable, and consequently, the probability that a job will run to completion [6]. If a machine fails while executing a job, the type of job will determine the effect of the failure.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…In other words, some nodes are completely dedicated to the grid; some provide service only when they are idle; and the rest falls between the two extremes [8]. Meanwhile, the resources can be of great heterogeneity, ranging from desktop PCs to supercomputers.…”
Section: Introductionmentioning
confidence: 99%
“…As mentioned before, system performance is vulnerable to the volatile donated grid resources. Grid nodes may become unavailable due to disconnection of power or communication, unwillingness or policies of the resource owners, etc., [8] describes a failure predictor, which utilizes historical data to forecast the availability of grid nodes. Thus, besides Makespan and Flowtime, it is important to take the reliability into account.…”
mentioning
confidence: 99%