Large Scale Backup Consolidation (No Deduplication) - HP 12000 Design Manual

Large Scale Backup Consolidation (No Deduplication)

Company requirements: Up to 500 TB backup storage required in a single unit, high throughput
required to meet backup windows.
Solution: Multi-node VLS9000 using Object copy to physical tape.
Caveats: Very large systems (greater than 300 TB) might be better split into two smaller systems
for increased availability. With a single large VLS, typically one storage pool is configured which
means if any full shelf in the VLS is lost (highly unlikely) then all virtual media is lost. But configuring
multiple storage pools to overcome this means increased manageability overhead.
ISVs: Data Protector, Netbackup, Networker, TSM, Commvault. See a full list at
www.hp.com/go/ebs.
Device configuration highlights: Create as many virtual devices as you need and dedicate devices
to certain SAN hosts by using the LUN mapping functionality.
Backup application configuration highlights: For best throughput, schedule as many backups in
parallel as you can. Use object copy functionality to schedule copy to physical tape.
Figure 26 Using VLS for Large Scale Consolidation
Recovery options: Typically direct from VLS, single file restores will be particularly fast; set retention
periods accordingly. Older copies of data can be recovered directly from physical tape if required.
Solution advantages:
Massive scalability in terms of capacity and performance, "hot add" capability for adding
capacity and nodes
High availability features in nodes and back-end disk arrays
Flexible emulations in terms of number of libraries, drives, slots, and media sizes
Easy disaster recovery solution if physical tape copy is regularly off-sited
Solution trade-offs: To guarantee scalable performance, each array module of 30 TB or 40 TB is
a single component and cannot be split. Sizing capacity and throughput correctly is important to
the solution being priced competitively.
66 VLS Devices
http://