Fileset considerations

You can create filesets based on conditions in your environment (for example, workflow patterns, security, or backup considerations, all the files used by a specific application, or files associated with a specific application or client). Filesets are used not only for managing the storage space used, but also for creating FlashCopy® images. Correctly defined filesets mean that you can take a FlashCopy image for all the files in a fileset together in a single operation, providing a consistent image for all of those files. The global namespace is partitioned into filesets that match the data-management model of the enterprise. Filesets can also be used as a criteria when placing individual files in global namespace.

Tip: When you are creating filesets, consider the overall I/O loads on the cluster. Because each fileset is assigned to one (and only one) metadata server, you need to balance the load across all metadata servers in the cluster by assigning filesets appropriately.

Separate filesets by their primary allegiance of the operating system to facilitate file sharing. Separating filesets also facilitates file-based backup methods (for example, utilities, such as tar, and Windows® backup applications such as VERITAS NetBackup or IBM® Tivoli® Storage Manager); full metadata attributes of Windows files can be backed up from a Windows backup client only and full metadata attributes of UNIX® files can be backed up from an UNIX backup client only.

There are no known limitations to the number of filesets that you can create; however, when the number of filesets is greater than one thousand, response time will increase when you issue fileset commands.

Parent topic: Filesets

Parent topic: How should I organize my global namespace?

Terms of use | Feedback
(C) Copyright IBM Corporation 2003, 2004. All Rights Reserved.