Search: onr:"swepub:oai:DiVA.org:ltu-14884" >
A Pretreatment Work...
A Pretreatment Workflow Scheduling Approach for Big Data Applications in Multi-cloud Environments
-
- Lin, Bing (author)
- College of Mathematics and Computer Sciences, Fuzhou University
-
- Guo, Wenzhong (author)
- College of Mathematics and Computer Sciences, Fuzhou University
-
- Xiong, Naixue (author)
- School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology
-
show more...
-
- Chen, Guolong (author)
- College of Mathematics and Computer Sciences, Fuzhou University
-
- Vasilakos, Athanasios (author)
- Luleå tekniska universitet,Datavetenskap
-
- Zhang, Hong (author)
- College of Mathematics and Computer Sciences, Fuzhou University
-
show less...
-
(creator_code:org_t)
- 2016
- 2016
- English.
-
In: IEEE Transactions on Network and Service Management. - 1932-4537. ; 13:3, s. 581-594
- Related links:
-
https://urn.kb.se/re...
-
show more...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- The rapid development of the latest distributed computing paradigm, i.e., cloud computing, generates a highly fragmented cloud market composed of numerous cloud providers and offers tremendous parallel computing ability to handle Big Data problems. One of the biggest challenges in Multi-clouds is efficient workflow scheduling. Although the workflow scheduling problem has been studied extensively, there are still very few primal works tailored for Multi-cloud environments. Moreover, the existing research works either fail to satisfy the Quality of Service (QoS) requirements, or do not consider some fundamental features of cloud computing such as heterogeneity and elasticity of computing resources. In this paper, a scheduling algorithm which is called Multi-Clouds Partial Critical Paths with Pretreatment (MCPCPP) for Big Data workflows in Multi-clouds is presented. This algorithm incorporates the concept of Partial Critical Paths, and aims to minimize the execution cost of workflow while satisfying the defined deadline constraint. Our approach takes into considerations the essential characteristics of Multi-clouds such as the charge per time interval, various instance types from different cloud providers as well as homogeneous intra-bandwidth vs. heterogeneous inter-bandwidth. Various types of workflows are used for evaluation purpose and our experimental results show that the MCPCPP is promising.
Subject headings
- NATURVETENSKAP -- Data- och informationsvetenskap -- Medieteknik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Media and Communication Technology (hsv//eng)
Keyword
- Distribuerade datorsystem
- Mobile and Pervasive Computing
Publication and Content Type
- ref (subject category)
- art (subject category)
Find in a library
To the university's database