Turtle: A Lan-Based Distributed Storage System

With the Internet exploding in size and reaching into every aspect of the society, digital data stored on-line are growing at an unprecedented rate. As a result, many organizations are under continuous pressure to expand their storage systems as demands for their services grow and their data sets swell relentlessly.Over the years the market has witnessed many technological innovations ranging from faster peripheral channel to dedicated storage area networks(SAN),finally to aggressively specialized storage systems using custom hardware and software. Unavoidably these highly specialized storage systems come with rising cost, which more often than not, makes many organizations fumble for more budget on storage systems. In recent years, the computer industry has made significant advances in magnetic recording technology. In the second half of 2006,the standard disk capacity on mainstream computers is about 160G and is growing continuously over time. However, the experiments show that in many organizations most of the local disk space on client workstations is only used for operating systems, application programs and temporary files, which in total take up only 4 to10GB disk space.Motivated by the increasing pervasive resource wasting ,this paper proposes a serverless distributed storage system Turtle utilizing idle disk space. For the convenience of users, Turtle provide a local filesystem interface, Meanwhile, with built-in data partition and redundant replica to achieve reliability and high availability.Turtle was not designed to replace the mainstream storage system. We just want to provide an alternative which is free. We don't promise that we have an excellent performance but we promise that we have a high reliability. This paper detailed described the design and implementation of Turtle include application environment, architecture, name space management, file organization, data dispatch and file system interface. Finally it analyzes the reliability of Turtle.