R&D: Providing Consistent State to Distributed Storage System

Computers has published an article written by Laskhmi Siva Rama Krishna Talluri, Department of Computer Science and Engineering, Koneru Lakshmaiah Educational Foundation, Vaddeswaram, Andhra Pradesh 522502, India, Ragunathan Thirumalaisamy, Department of Computer Science and Engineering, SRM University-AP, Andhra Pradesh 522502, India, Ramgopal Kota, Department of Physical Layer Products, Broadcom, Hyderabad 500032, India, Ram Prasad Reddy Sadi, Department of Information Technology, Anil Neerukonda Institute of Technology & Sciences, Visakhapatnam, Andhra Pradesh 531162, India, Ujjwal KC, Ranesh Kumar Naha, School of Information and Communication Technology, University of Tasmania, Hobart, TAS 7005, Australia, and Aniket Mahanti, School of Computer Science, University of Auckland, Auckland 1010, New Zealand, and Department of Computer Science, University of New Brunswick, Saint John, NB E2L 4L5, Canada.

Abstract: “In cloud storage systems, users must be able to shut down the application when not in use and restart it from the last consistent state when required. BlobSeer is a data storage application, specially designed for distributed systems, that was built as an alternative solution for the existing popular open-source storage system-Hadoop Distributed File System (HDFS). In a cloud model, all the components need to stop and restart from a consistent state when the user requires it. One of the limitations of BlobSeer DFS is the possibility of data loss when the system restarts. As such, it is important to provide a consistent start and stop state to BlobSeer components when used in a Cloud environment to prevent any data loss. In this paper, we investigate the possibility of BlobSeer providing a consistent state distributed data storage system with the integration of checkpointing restart functionality. To demonstrate the availability of a consistent state, we set up a cluster with multiple machines and deploy BlobSeer entities with checkpointing functionality on various machines. We consider uncoordinated checkpoint algorithms for their associated benefits over other alternatives while integrating the functionality to various BlobSeer components such as the Version Manager (VM) and the Data Provider. The experimental results show that with the integration of the checkpointing functionality, a consistent state can be ensured for a distributed storage system even when the system restarts, preventing any possible data loss after the system has encountered various system errors and failures.“