Abstract | ||
---|---|---|
With the rapid grows of cloud-based internet application,a need for efficient resource allocation, load balanceand cost management increases. In this paper, we propose agroup-auction based mechanism for the cloud instance marketto efficiently allocate resources. In the market system, resourceproviders offer resources in the form of virtual machine. Userssubmit their bids. The proposed system assists to decide howproviders allocate their resources to users efficiently. Groupauction is also used to reduce cost and improve sales volumefor the participants. To find how to allocate resources, the utilityoptimization problem is formulated as a Distributed ConstraintOptimization Problem (DCOP), and we apply a distributedmessage passing algorithm, max-sum algorithm using the groupauction formation game to find the nearly optimal solution withquadratic time complexity. In comparison study, we deploythe system on two scenarios. In terms of resource efficiencyand acceptance rate, we compare our algorithm with greedyalgorithm and random algorithm. Experiment results show thatour system performs very well in probable setting. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/.50 | CBD |
Keywords | Field | DocType |
algorithm,PCDN,search and routing | Resource management,Randomized algorithm,Algorithm design,Computer science,Greedy algorithm,Resource allocation,Time complexity,Auction algorithm,Distributed computing,Cloud computing | Conference |
ISSN | Citations | PageRank |
1066-6192 | 1 | 0.36 |
References | Authors | |
9 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yiyi Ma | 1 | 1 | 0.36 |
Bin Li | 2 | 68 | 27.40 |
Yonglong Zhang | 3 | 9 | 4.91 |
Junwu Zhu | 4 | 72 | 16.11 |