Efficient Load Balancing Using Active Replica Management in a Storage System

Jui-Pin Yang

Mathematical Problems in Engineering, 2016, vol. 2016, 1-9

Abstract:

Many algorithms can uniformly distribute data to storage nodes in a storage system. However, it cannot avoid load imbalance because data has different popularity. To resolve this issue, we propose a novel dynamic replication scheme, namely, Active Replica Management (ARM). ARM actively establishes optimal number of copies for hotspot data according to data access behaviors and then efficiently distributes copies to other storage nodes based on current amount of copies related to hotspot data. To improve storage utilization, ARM automatically and gradually dereplicates the useless copies of hotspot data when they become nonhotspot data. ARM resolves load imbalance by allocating dynamic copies to adequate storage nodes, and hence it can prevent partial storage nodes from overburdening. Simulation results demonstrate that ARM is an efficient scheme with excellent performance on load balancing, significantly closer to Optimal Load Balancing (OLB). In addition, ARM’s performance outperforms both Static Load Balancing (SLB) and No Replica schemes.

Date: 2016
References: Add references at CitEc
Citations:

Downloads: (external link)
http://downloads.hindawi.com/journals/MPE/2016/4751829.pdf (application/pdf)
http://downloads.hindawi.com/journals/MPE/2016/4751829.xml (text/xml)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hin:jnlmpe:4751829

DOI: 10.1155/2016/4751829

Access Statistics for this article

More articles in Mathematical Problems in Engineering from Hindawi
Bibliographic data for series maintained by Mohamed Abdelhakeem ().