Hitachi Unified Storage VM, Руководство по обзору продукта, Страница: 43 / 70

Software overview
3–7
Hitachi Unified Storage VM Product Overview Guide
• Data relocation: Data may be relocated from any given file system to
another using a mechanism referred to as transfer of primary access
(TPA). TPA makes it possible to relocate individual directories as well as
entire file systems. TPA does, however, involve a small amount of
downtime, and data is no longer accessible using the same path names
that were in use prior to the relocation. TPA is generally used to better
organize file systems and/or directories within them.
Data Protection
The Hitachi File System also offers features for online data recovery, data
replication, mirroring, backup, disaster recovery, and complete system
monitoring.
Hitachi NAS solutions support high availability (HA) clustering of servers in
a 2-node active-active configuration or an n-way (more than 2 nodes)
cluster configuration. Clustered servers provide NVRAM mirroring for
enhanced data protection, automated file system failover, and higher levels
of performance as additional servers are added to the cluster.
In addition, the Hitachi File System provides additional mechanisms for data
protection. Three of the more important mechanisms are snapshots, data
replication, and data backup. Snapshots are generally described as point-
in-time copies of the file system, and are a very convenient way to give end-
users a way to "roll back" to a previous point in time to recover their own
data. There are several data replication options within the Hitachi NAS
system software; these may be described as either file, object or block-
based, and synchronous or asynchronous.
High availability
The Hitachi File System high availability (HA) design is the concept of
enterprise virtual servers (EVS). Virtual servers are logical entities that
reside on a physical server, similar to operating system virtualization
techniques such as VMware, Microsoft Hyper-V, or Citrix XenServer. An EVS
does not have physical interfaces, but instead has virtual interfaces that
map to the physical interfaces of the server. As a result of the separation
between the physical and logical interfaces of the EVS and the physical
server, an EVS can be migrated from one physical server to another,
transparently. In a failover situation for an HA cluster, automated EVS
migration process takes place without system shutdown, and in most cases
occurs quickly enough that hosts using stateless protocols (such as NFSv3)
will not require unmounting and remounting of NFS exports.
Snapshots
Snapshots allow the storage administrator to capture a point-in-time image
of the file system and the point-in-time image is a read-only view of the file
system. Using point-in-time images (snapshots), the storage administrator
can:
• Allow end-users to retrieve files that have been deleted without
administrator intervention