Puredisk CR Queue Processing - backups run MUCH faster when this is running!

[gc_bus]

We have two Puredisk SPA's and each consists of an unclustered setup of two content routers with one being the metabase engine as well.

We use Netbackup via the PDDO plugin to write to the primary Puredisk SPA and replicate to an off-site SPA via Netbackup Storage Lifecycle Policies. (optimised duplication).

We currently have had occassion to write directly to the off-site SPA whilst we upgrade the primrary SPA.

I have noticed that the Netbackup PDDO backups to the off-site SPA only run at acceptable speeds when the off-site SPA is running Content Router Queue Processing jobs. When they have completed then the backup performance takes a nose-dive. I would have expected the opposite to have been true.

A case has been raised with Symantec but just wondering if anyone else has seen this.

Cheers.

[gc_bus · Accepted solution]

Fixed by increasing the content routers Max Cache setting (via the Web gui) to a value of 65% so that all of the fingerprints fit into memory. Needed a Puredisk services restart.