GlusterFS
NAS Recovery - GlusterFS
GlusterFS is a network/clustered file system offering scalability up to several petabytes. It supports different types of communication transports such as TCP/IP, InfiniBand VAPI/Verbs, Sockets Direct Protocol and UNIX domain sockets. It aggregates various storage bricks over these interconnects into a single large parallel network file system.
GlusterFS is based on a stackable user space design. It consists of two components: a server and a client. The storage server runs glusterfsd and the clients use mount command or glusterfs client to mount the exported filesystem.
To mount GlusterFS file systems, clients need Filesystem in Userspace (FUSE) support in the kernel. Servers can run on any node. GlusterFS server is tested on Linux, FreeBSD and Opensolaris, and client runs on only Linux machines. From 1.3.9 release onwards GlusterFS runs both client and server on Mac OS X Leopard. There are no minimum system requirements as such.
Advantages
- Since GlusterFS aggregates on top of existing filesystems, users can recover the files and folders even without GlusterFS.
- GlusterFS is completely distributed. It has no single point of failure.
- GlusterFS has a highly modular design, extensible through powerful translator mechanism.
- Automatic File Replication.
- Can use Stripe translator for getting more I/O performance for large files.
- Since it does not require kernel patches, the software can be installed without any server downtime.
- Implemented entirely in user space. Easy to port, debug and maintain.
- Highly scalable.
Unlike other cluster file systems which addressed the parallelization at block level, GlusterFS addresses the problem at volume management and I/O scheduling level.
