On the Dynamic Maintenance of Data Replicas based on Access Patterns in A Multi-Cloud Environment

Abstract

Abstract-Cloudcomputing provides services and infrastructures to enable end-users to access, modify and share massive geographically distributed data. There are increasing interests in developing data-intensive (big data) applications in this computing environment that need to access huge datasets. Accessing such data in an efficient way is deterred with factors such as dynamic changes in resource availability, provision of diverse service quality by different cloud providers. Data replication has already been proven to be an effective technique to overcome these challenges. Replication offers reduced response time in data access, higher data availability and improved system load balancing. Once the replicas are created in a multi-cloud environment, it is of utmost importance to continuously support maintenance of these replicas dynamically. This is to ensure that replicas are located in optimal data center locations to minimize replication cost and to meet specific user and system requirements. First, this paper proposes a novel approach to distributed placement of static replicas in appropriate data center locations. Secondly, motivated by the fact that a multi-cloud environment is highly dynamic, the paper presents a dynamic replica maintenance technique that reallocates replicas to new data center locations upon significant performance degradation. The evaluation results demonstrate the effectiveness of the presented dynamic maintenance technique with static placement decisions in a multi-cloud environment.

Review on data replication strategies in single vs. interconnected cloud systems: Focus on data correlation‐aware strategies

“…In order to provide the tenants' tasks with the required QoS, these resources should have sufficient storage space and acceptable bandwidth with other resources that require replicas. Moreover, they should not be overloaded either 84,115 Replicas number adjustment consists of determining how many replicas to maintain in order to meet the required QoS.…”

“…Moreover, they should not be overloaded either. 84,115 • Replicas number adjustment consists of determining how many replicas to maintain in order to meet the required QoS. This process also includes selecting the replica to be removed or replaced.…”

“…27,59,63,128 The reduction of network and storage usage is widely addressed by the strategies. 27,60,76,79,115 It is usually performed by maintaining a minimum number of replicas. 75,116 Other strategies assess the cost of replication as a function of time.…”

Replication Strategy with Comprehensive Data Center Selection Method in燙loud Environments

“…Availability of more space in storage gives higher opportunities for the data center to be chosen as the best candidate for replica storage [45][46][47][48][49][50][51]. Consequently, Storage merit accumulation in this research is to identify available storage space in each DC x .…”

Leveraging A Multi-Objective Approach to Data Replication in Cloud Computing Environment to Support Big Data Applications

“…The simulation results demonstrate an important reduction in terms of number of replicas that violate the QoS requirement in contrast to the existing replication strategy adopted by Hadoop. In a recent effort, Shorfuzzaman [35] presents a www.ijacsa.thesai.org dynamic replica maintenance technique called DRMS in a multi-cloud environment taking into account various QoS requirements from users. Periodically replicas are relocated to new locations upon significant performance degradation.…”