Certain investigations on storing and retrieving files in open source cloud

Abstract

In the new era of information technology, Cloud computing brings a great challenge in storing and retrieving more user files in remote locations. Future file system needs exabytes of data to store trillions of user files. This scalability in storage becomes a bottleneck in storing, organizing and retrieving files in a cloud system. However, providing effective search, store and organize the trillions of files using the existing algorithms is not a simple task. newlineIn the cloud storage system, more user files are stored on different compute nodes. To organize and maintain files in the cloud storage system, digit binate algorithm and digit compact prefix algorithms are used in the backend of the compute node and controller node. The algorithm time complexity is analysed and discussed. The performance evaluation done for both digit binate and digit compact prefix algorithms. The implementation is done in OpenStack private cloud. The digit binate algorithm shows the minimum time in storing and retrieving compared to digit binate algorithm with different compute node instances and files. The implementation is done in OpenStack private cloud. newline