User's Manual

262
partition
(FSE partition)
A configurable entity used by the FSE implementation to split the storage and configure it
according to user requirements. Each FSE partition is related to a single HSM file system and
one or more FSE media pools. It has its own policies, File System Catalog, and Hierarchical
Storage Management Database.
Partition Manager
(PM)
An FSE daemon/service that executes and controls the migration, release, and recall processes
for a single FSE partition, according to the policies defined in the FSE partition configuration
file. The Partition Manager process name is fse-pm.
performance disk A disk with high-performance characteristics that stores HSM file systems. Performance disks
must be located on the systems that host FSE clients.
primary storage
space
A logical volume/disk partition used to store the HSM file system.
process
(FSE process)
A process running on an FSE implementation that performs FSE-specific tasks.
recall policy A rule by which recall jobs for files on an HSM file system are conducted. Since recall is started
on an FSE user's request, the only configurable parameter is one that specifies the maximum
time a user should wait for the file to become online. This parameter is defined in the FSE
partition configuration file. Each FSE partition has its own recall policy.
recall retention time A period of time during which a recalled file is kept online before it is released again. See also
”retention time”.
recall, recall job A process, executed by the Partition Manager, that copies the file contents from the FSE media
to the HSM file system. Once the process is completed, the offline file is again online. Its file
entry is then added to the release candidate list.
recovery
(FSE recovery)
A process, executed by the Partition Manager, that recovers the HSM file system or the File
System Catalog (FSC) without requiring FSE backup copies. The HSM file system is recreated
from the Name Space Catalog (NSC), while the FSC is recreated either from Fast Recovery
Information (FRI) or from migrated metadata. FSE recovery can be used as a fallback if restore
of FSE is not possible due to missing FSE backup data. A recovery job is triggered explicitly on
an FSE user's request with the fserecover command. See also ”recovery (FSE recovery)”.
recovery job See ”recovery (FSE recovery)”.
recycled medium
volume
An FSE medium volume from which all selected data was copied to another medium volume
during media reorganization. The data on recycled medium volumes can no longer be
retrieved, but the storage space on such volumes can be reused for new migrated data.
redundant data copy
recreation
A process in FSE that recreates a copy of a file set that has become unreadable, most probably
because the FSE medium to which it was migrated is damaged. The process recreates the
damaged data copy using copies of the same data that were migrated to other FSE media
pools. Therefore, the main prerequisite to execute this process is a multiple media pool
configuration of the corresponding FSE partition. A redundant data copy recreation process is
triggered manually by the FSE administrator.
regular release Release of files on an HSM file system that are on the release candidate list and meet the
predefined criteria, triggered when the high watermark of usage is reached on the HSM file
system. The process is stopped when the low watermark on the HSM file system is reached. See
also ”release”.
release A process that usually follows a migration or recall, executed by the Hierarchical Storage
Manager. This process removes the file data from the HSM file system, leaving only the file
header. Release is triggered by the Partition Manager according to the defined watermarks.
After release, files are offline and their file entries are removed from the release candidate list.