Increased data availability

GPFS provides multiple features that improve the reliability of your file system. This includes automatic features like file system logging and configurable features like intelligently mounting file systems on startup to providing tools for flexible synchronous replication.

GPFS allows you to organize your storage hardware into failure groups. A failure group is defined as a set of disks that share a common point of failure that could cause them all to become simultaneously unavailable. Failure groups are defined by the system administrator, so care needs to be taken when defining disks to ensure proper failure group isolation. When used in conjunction with the replication feature of GPFS, the creation of multiple failure groups provides for increased file availability should a group of disks fail. Replication in GPFS ensures that there is a copy of each block of replicated data and metadata on disks in different failure groups. In this case, should a set of disks become unavailable, GPFS fails over to the replicated copies in another failure group.

During configuration, you assign a replication factor to indicate the total number of copies of data and metadata you wish to store. Currently, the maximum replication factor is 3. Replication allows you to set different levels of protection for each file or one level for an entire file system. Since replication uses additional disk space and requires extra write time, you should consider the impact of replication on your application, especially if the replication is occurring over a WAN. To reduce the overhead involved with the replication of data, you may also choose to replicate only metadata as a means of providing additional file system protection. For further information on GPFS replication, see File system replication parameters.

GPFS is a logging file system. It creates separate logs for each file system. GPFS automatically replicates recovery logs if multiple failure groups are available. When used in conjunction with geographic based replication, disaster recovery abilities are provided. For further information on failure groups, see Network Shared Disk (NSD) creation considerations. For further information on disaster recovery with GPFS, see Data Mirroring and Replication.

Once your file system is created, it can be configured to mount whenever the GPFS daemon is started. This feature assures that whenever the system and disks are up, the file system will be available. When utilizing shared file system access among GPFS clusters, to reduce overall GPFS control traffic you may decide to mount the file system when it is first accessed. This is done through either the mmremotefs command or the mmchfs command using the -A automount option. GPFS mount traffic may be reduced by using automatic mounts instead of mounting at GPFS startup. Automatic mounts only produce additional control traffic at the point that the file system is first used by an application or user. Mounting at GPFS startup on the other hand produces additional control traffic at every GPFS startup. Thus startup of hundreds of nodes at once may be better served by using automatic mounts. However, when exporting the file system through Network File System (NFS) mounts, it might be useful to mount the file system when GPFS is started.