An Approach About Cloud-of-Clouds Storage Scaling Based On RAID-5

Recently,with the rapid growth of data,traditional storage scheme has been difficult to meet the user's storage requirements.As a new distributed storage mode,cloud storage is increasingly favored by researchers and users.Cloud-of-clouds storage has gradually become a focus of scientific research areas since it solves the problem of a single point of failure in single cloud storage and mitigated vendor lock-in.The Cloud-of-clouds storage scaling problem is an important part of it.Due to the need for a large amount of bandwidth while scaling,and commercial cloud bandwidth resources are more expensive,therefore,how to reduce the scaling bandwidth as much as possible has become the focus of the study.In this paper,cloud computing services and object storage services are combined to build distributed cloudy storage architecture.The object storage service is used for data storage,and the corresponding computing service is used to calculate the stored data.The cloudy platform's data distribution is RAID-5.In the case where the total amount of data and data distribution are unchanged,designing and implementing a scaling algorithm which exploits the clouds' internal cheap computing resources to calculate the internal data and then transfer.The algorithm reduces the scaling bandwidth by decentralizing the computing tasks.It can be seen that the number of blocks that need to be transmitted is exactly the number of blocks required by the new cloud,so the data transmission is optimal.In order to further improve the efficiency of scaling,this paper uses multithreading and other measures to optimize the scaling process,while ensuring the cloud load balancing while scaling.The experimental results show that this method significantly reduces the cost of scaling and improves the scaling efficiency compared with the RR method and the GSR method.The results show that this RAID-5 scaling method has the least cost,minimal bandwidth,and significantly reduced scaling time in cloud-of-clouds storage system.