Bulut depolama sistemleri için sistem erişilebilirliğini artıran veri kopyalama stratejisi

The Cloud environment constitutes a heterogeneous and a highly dynamic environment. The Cloud Computing provides the software and hardware infrastructure as services using large-scale data centers. As a result, Cloud Computing moved away the computation and data storage from the end user and onto servers located in data centers, thereby relieving users of the burdens of application provisioning and management and enabling them to focus on managing of their application logic. However, failures on the data center storage nodes often take place in cloud computing environments. As a result, the cloud environment requires some capability for an adaptive data replication management in order to cope with the inherent characteristic of the Cloud environment. To improve system availability and get high fault tolerance, replicating data is a good choice. Replication is the process of providing different replicas of the same service at different nodes. In most of the real cloud, data replication is achieved through data resource pool, the number of data replicas is statistically set based on history experience and is usually less than 3. This strategy works well at most time, but it will fail at inclement times. Replicating the data with a fixed number of its copies to multiple suitable locations should not be an advisable choice. How to decide a reasonable number and right/suitable locations for replicas has become a challenge in the cloud computing. In this thesis, we present a new dynamic data replication strategy suitable for cloud environments. Our technique provides optimum replica number and prevents the replica number from incrementing in time. In addition, it enables the replicas to be placed to all data center nodes in a balanced way depending on the available disk space for a disk, expected availability of requesting cloud users, failure probability of each node in data center servers. We also conclude that bandwidth usage can be reduced by considering the distance between user and data center servers and locating the replicas nearby users who requests for.