S3QL is a file system that uses something like Google Storage, http://www.aliyun.com/zixun/aggregation/2467.html ">amazon S3 or OpenStack storage devices to store all of the online data. It supports Amazon S3, as well as any SFTP server, and effectively provides a dynamic and infinitely capable hard drive that can be accessed from any computer that has access to the Internet. S3QL provides a standard, full-featured UNIX file system that is conceptually indistinguishable from any local file system. S3QL has additional features such as compression, encryption, duplicate data deletion, tree-like structure and snapshots of invariant patterns, making it particularly suitable for online backup and archiving. The main purpose of S3QL is to provide users with simple and elegant performance as well as flexible features.
S3QL 1.7 This version of the metadata can be customized for storage, the delta-encoded binary format and Inode are assigned sequentially and not reused, making S3QL fully compatible with NFS. S3QLCP you can now copy extended properties and no longer produce incorrect block reference counts, which have the same block when you copy a file. This version removes the S3QLCP race condition. When a file is copied, immediately modified or a S3QLCP is created, it is possible that S3QLCP will replicate the new file properties.
Software Information: http://code.google.com/p/s3ql/
Download Address: http://code.google.com/p/s3ql/downloads/detail?name=s3ql-1.7.tar.bz2&can=2&q=