Replication is one of the important roles in cloud storage to improve data availability, fault tolerance and throughput for users and control storage cost. As data access pattern changes every time, the nature of popular files is unpredictable and unstable. Therefore, data popularity is taken into account as an important factor in replication. Data popularity in replication impacts an efficient storage because it is able to reduce waste storage for unpopular files. Also, data locality is a key issue in storage system and this consequence occurs performance overhead of system. Therefore, this paper introduces a replication strategy for cloud storage. The proposed strategy contains two portions; replica popularity and replica placement. First for replica popularity, popularity is taken into account by analyzing the changes in data access pattern. Second for replica placement, replicas are placed and performed on dedicated assigned nodes in order to enhance data locality. The proposed placement algorithm is able to avoid the overloaded problem of nodes by considering the load of nodes such as disk utilization, adjustable disk bandwidth and CPU utilization. This proposed strategy will be efficient for cloud storage.
Popularity, Data Locality, Disk Utilization, Adjustable Disk Bandwidth, CPU Utilization
May Phyo Thu , Khine Moe Nwe , Kyar Nyo Aye "Data Popularity - Aware Replication Strategy For Cloud Storage" Iconic Research And Engineering Journals Volume 3 Issue 2 2019 Page 494-500
May Phyo Thu , Khine Moe Nwe , Kyar Nyo Aye "Data Popularity - Aware Replication Strategy For Cloud Storage" Iconic Research And Engineering Journals, 3(2)