2007
DOI: 10.1109/tpds.2007.1094
|View full text |Cite
|
Sign up to set email alerts
|

Adaptive Reputation-Based Scheduling on Unreliable Distributed Infrastructures

Abstract: This paper addresses the inherent unreliability and instability of worker nodes in large-scale donationbased distributed infrastructures such as P2P and Grid systems. We present adaptive scheduling techniques that can mitigate this uncertainty and significantly outperform current approaches. In this work, we consider nodes that execute tasks via donated computational resources and may behave erratically or maliciously. We present a model in which reliability is not a binary property but a statistical one based… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
64
0
1

Year Published

2008
2008
2017
2017

Publication Types

Select...
4
3
2

Relationship

1
8

Authors

Journals

citations
Cited by 76 publications
(66 citation statements)
references
References 29 publications
0
64
0
1
Order By: Relevance
“…authors in [13] have defined an interesting reputation formula which addresses the inherent unreliability and instability of worker nodes in large-scale donation based distributed infrastructures such as P2P and Grid systems. In the paper authors present adaptive r -reputation, completed_tasks -number of tasks completed successfully, all_tasks -number of all tasks in the system.…”
Section: Related Workmentioning
confidence: 99%
“…authors in [13] have defined an interesting reputation formula which addresses the inherent unreliability and instability of worker nodes in large-scale donation based distributed infrastructures such as P2P and Grid systems. In the paper authors present adaptive r -reputation, completed_tasks -number of tasks completed successfully, all_tasks -number of all tasks in the system.…”
Section: Related Workmentioning
confidence: 99%
“…A most recent proposal or reputation-driven scheduling in context of voluntary computing environments (desktop grids) has been put forward by Jason et al [17]. They consider a centralized system model, where a central server is assigned responsibility for maintaining reliability ratings that form the basis for assigning tasks to group of voluntary nodes.…”
Section: Dependable Schedulingmentioning
confidence: 99%
“…Uncertainty and unreliability are facts in federated environments, which are triggered by multiple factors, including: (i) software and hardware failures as the system and application scale that lead to severe performance degradation and critical information loss; (ii) dynamism (unexpected failure) that occurs due to temporal behaviours, which should be detected and resolved at runtime to cope with changing conditions; and (iii) lack of complete global knowledge that hampers efficient decision making as regards to composition and deployment of the application elements. The existing Grid scheduling approaches and application composition techniques [12], [17] are inadequate to handle all of these uncertainties caused by the dynamic behaviours related to resources and applications in federated environments.…”
Section: Introductionmentioning
confidence: 99%
“…In addition to the above methods, majority-voting methods [13] have been used, such as correlation-and reputation-based methods. Correlation-based methods use the results of votes to partition honest workers from colluders [14].…”
Section: Introductionmentioning
confidence: 99%