Testing Backups To Tape Vs. Backups To A Vls - HP StorageWorks 12000 - Virtual Library System EVA Gateway Manual

Figure 49 Testing Backups to Tape vs. Backups to a VLS
.
If you run two tests, in both cases backing up approximately 13 GB of data from the five servers, the
following results:
• Backing up to physical tape takes 6 minutes 41 seconds
• Backing up the same data to the VLS takes 3 minutes 34 seconds
Example 3: VLS12000 EVA Sizing Without Deduplication
When sizing the EVA array requirements for a VLS12000 without deduplication enabled, the key
factor to take into account is the compressibility of the data since the hardware compression built into
the VLS12000 nodes will reduce the data rate going to the EVA arrays by the compression factor.
For example, a VLS12000 with 5 nodes has a 3000 MB/s backup performance capability. With
2:1 data compressibility this would only require 1500 MB/sec of EVA array performance. Therefore,
you would need at least three EVA 4400/6xxx/8xxx arrays (each array with 8 shelves full of disks)
to meet that requirement.
Example 4: VLS12000 EVA Sizing With Deduplication
When sizing the EVA array requirements for a VLS12000 with deduplication enabled, first calculate
the base array performance required without deduplication (see above) and then for deduplication
you double this performance to get the required array performance. This is because the VLS
deduplication is post-process and therefore will add additional disk I/O requirements on top of the
backup I/O. For example, a VLS12000 with 3 nodes and deduplication has a 1500 MB/s backup
performance capability. With 2:1 data compressibility this would only require 750MB/sec of base
EVA array performance plus another 750MB/sec for the deduplication, giving a total of 1500MB/sec
array performance. Therefore, you would need at least six EVA 4400/6xxx/8xxx arrays (each array
with 8 shelves full of disks) to meet that requirement.
126 Virtual Library Systems