How does Swift-Open Stack Storage System manage file metadata?

Question

I am working on writing a paper on the cloud storage.

I am looking into Openstack storage system called Swift.

On the page: http://swift.openstack.org/overview_architecture.html

Object Server The Object Server is a very simple blob storage server that can store, retrieve and delete objects stored on local devices. Objects are stored as binary files on the filesystem with metadata stored in the file’s extended attributes (xattrs). This requires that the underlying filesystem choice for object servers support xattrs on files. Some filesystems, like ext3, have xattrs turned off by default.

Does it mean Swift does not separate the metadata from the file content? Would it create bottlenecks on huge number of metadata access or search? With each Head Object access, would it go through the file system to find the metadata?

Thanks

Answer 1

The swift takes advantages of md5 hash and its own ring data structures to look up a file's path. When it locates a file, then it reads the xattrs with the file. As a result, it does not need to read the all metadata in filesystem.

Answer 2

This book can help you a bit in understanding swift. However this book does not provide the answer for your question but can help you thinking how it is stored. For better understanding , you necessarily need to hack the swift code for yourself and see or approach the swift emailing list i.e. swift@lists.launchpad.net. Also find the following link Swift Api docs