Clone-based Data Index in Cloud Storage Systems

Abstract

Abstract. The storage systems have been challenged by the development of cloud computing. The traditional data index cannot satisfy the requirements of cloud computing because of the huge index volumes and quick response time. Meanwhile, because of the increasing size of data index and its dynamic characteristics, the previous ways, which rebuilding the index or fully backup the index before the data has changed, cannot satisfy the need of today`s big data index. To solve these problems, we propose a double-layer index structure that overcomes the throughput limitation of single point server. Then, a clone based B+ tree structure is proposed to achieve high performance and adapt dynamic environment. The experimental results show that our clone-based solution has high efficiency.

File retrieval and storage in the open source cloud tool using digital bipartite and digit compact prefix indexing method

“…Each data center node maintains a part of the global index. This type of storage and retrieval is impossible in large cloud systems …”