PD SPA Replicated data showing markedly different reduction factor

[Kev_D]

Hi all.

We have a NetBackup 7.0.1 Master Server controlling SLP driven backups to 2 Puredisk 6.6.1 SPAs.

The curious thing is that:

• on the SPA where the primary images are housed, a data mining report shows that about 30Tb of images are being protected by using 6.5Tb of CR space
• on the SPA where the duplications are housed the mining report shows a similar amount (30Tb) of image data is being held using only 1.9Tb!!

I would expect the space consumed to be similar.

I ran a crcontrol –dsstat on both to see if there was perhaps a large amount of space being reserved in “all but empty” containers on the primary SPA, but that does not seem to be the case.

Is this expected behavior, or can any of you shed any light on why the duplication SPA seems so much more efficient?

thanks!

[gc_bus]

I suspect that on your NBU master/media server(s) that you have not got compression set on (look at the pd.conf file) and if you use client side de-dupes you need to look at them too. No compression is the default for NBU to PD backups but for SLP dupes, compression is the default. Hence the difference in actual PD space use between your two SPA's.