EMC Networker and Replication WP LDV
EMC Networker and Replication WP LDV
Abstract
IT departments are under pressure to deliver higher levels of
backup and recovery services than ever before to keep pace with
round-the-clock business operations. By combining
management of heterogeneous replication and snapshot
technologies with backup operations, EMC® NetWorker®
provides a unified “command-and-control” platform to deliver
unparalleled performance and reduced complexity for mission-
critical application and data protection.
February 2011
Copyright © 2007, 2008, 2009, 2011 EMC Corporation. All
Rights Reserved.
For the most up-to-date listing of EMC product names, see EMC
Corporation Trademarks on EMC.com.
Audience
This white paper is intended for anyone who is using, considering, or selling
NetWorker.
Split-mirrors
Split-mirror, also referred to as disk mirroring, is a replication technique that
duplicates every byte of the original data volume to another volume. These copies are
known as business continuance volumes (BCVs), mirrors, and clones. A mirror can be
temporarily suspended or split to create a point-in-time copy of data. During a split,
the disk subsystem temporarily stops making updates to the mirror copy, which
allows a frozen data point. The split-mirror can then be mounted read/write or used
for backup and recovery. A recovery will impact the production volume; all other
activities do not impact the production volume. After an offline backup is complete,
the mirror is established and resynchronized with the product volume. A full data
copy is available within the mirrored copy and therefore requires 100 percent of the
capacity of the source. For example, 1 TB of data requires 1 TB of disk space for a
mirror copy. With a mirror, if the original volume is lost, the alternate volume is an
exact copy of the original. The issue of disk space must be considered if the intent is
to store multiple copies.
Production view
of volume Original
Snapshot view
of volume Mirror
Copy-on-write
Copy-on-write replication, also called snaps or snapshots, creates a virtual copy of an
original volume. When the snapshot is initiated, the file system is frozen and a cache
of disk space is created. Writes made after the initiation of the snapshot trigger a
copy of the original block(s) to the cache. The production disk contains all current
data, while the snap cache contains any original data that has subsequently been
altered. A snapshot can be mounted for read/write access, or used for backup and
recovery of the production volume. To recover the original disk to a point in time, the
data from the snap cache is moved back to the original to re-create the volume as it
existed at the time the snapshot was taken. Consequently, the copy-on-write
snapshot is dependent on the original volume and a mount or recovery operation
from the snapshot must hit the production disk. The required disk cache size will vary
depending on the rate of change as well as the frequency and retention period of
snapshots. Typically copy-on-write requires far less space than a mirror – on average
10 percent to 20 percent of the source volume size for each snapshot. Space
requirements depend on how many writes and changes are made to the source
volume and how long the snapshot is active.
Snapshot view
Cache of volume
PowerSnap Modules
PowerSnap Modules coordinate with NetWorker application support for Oracle, SAP,
SQL Server, and IBM DB2 to create consistent replicas of messaging and database
applications that reside on the supported storage technologies. Similar to NMM, once
a point-in-time replica is created, PowerSnap verifies that the copy is clean and
mountable to ensure recoverability and to enable off-host backup and restore
operations to be performed as necessary. Snapshot policies are established and
assigned in the NetWorker’s management interface. In addition, PowerSnap provides
command line utilities to enable the browsing and management of replicas. From
these utilities administrators can accomplish the following:
• Recover whole snapshots
• Recover individual files and directories from snapshots
• Generate diagnostic reports
When recovery of an application from a snapshot is required, the restore process is
managed from the command line utilities or the PowerSnap-enabled Application
Module interface.
Snapshot-assisted backup
There are several backup types available when using NetWorker in conjunction with
snapshots.
Instant backup
A point-in-time copy of data that is initiated and stored as a snapshot session or
instance is called an instant backup. An instant backup is a block-level snapshot
created from the application server and made available on disk but not written to
tape. The snapshot is registered within the NetWorker media database to facilitate
tracking for recovery.
Proxy backup
To lessen the impact on file and application servers a proxy client or proxy host can
mount and back up a volume or file system belonging to another client. The proxy
client moves the actual data created by the application server within the snapshot or
mirror to the backup storage such as tape or disk. While the proxy host is a server,
this type of backup is referred to as serverless because it does not require the original
application host to facilitate data movement. Figure 4 illustrates proxy host backup to
disk or tape.
Snapshot-assisted recovery
There are several PowerSnap restore types.
Rollback
A rollback is the process of returning data to an earlier point-in-time copy in response
to a recovery operation, and is a complete restore from a point-in-time copy to a
standard volume without host involvement. Rollback restores are destructive by
nature and can occur without having to retrieve data from the secondary storage
(tape/disk).
Rollforward
A rollforward is the process of progressing data from a rollback using one or more
instant backups. For example, if three snapshots were created at 10 A.M., 11 A.M.,
and 12 P.M., the user can perform a rollback to the 10 A.M. snapshot and then a
rollforward to the 11 A.M. snapshot or even the 12 P.M. snapshot. You may perform a
rollback from a more recent copy to approximate the same effect.
File-level recovery
When NetWorker initiates an instant backup, it captures an index to enable file-level
recovery from a mounted replica. The recovery process is also referred to as instant
recovery or file-by-file recovery.
Conclusion
EMC has a complete recovery management vision supported by the most robust
portfolio of recovery management products on the market, which meet every
requirement of a deployment framework. NetWorker and its options for snapshot
management deliver on the promise of combining protection technologies to provide
higher degrees of protection and recoverability for business data and information
assets directly mapped to an organization's RPOs and RTOs.