Discussions specific to the VMware vSphere hypervisor
Post Reply
Posts: 97
Liked: 2 times
Joined: Feb 21, 2020 11:30 am
Full Name: massimiliano porliod

data domain primary repository

Post by mporliod »

Hello everyone, I have this curiosity is it true that Veeam does not recommend using the data domain as the primary repository because data recovery is very slow due to deduplication?
Is this still true or has it been overcome.

Andreas Neufert
VP, Product Management
Posts: 6358
Liked: 1330 times
Joined: May 04, 2011 8:36 am
Full Name: Andreas Neufert
Location: Germany

Re: data domain primary repository

Post by Andreas Neufert »

Veeams best practices is to store the primary backups on (fast) general storage to optimize backup and restore. We build our functionality with this concept in mind, so everything is optimized to leverage random IO processing.
Usually you hold data there for some time (like 14days) and create backup copies to a more cost effective storage. This can be a deduplication storage for long term retention.
Deduplication storages are optimized for sequential processing and can be used (supported) as primary storage as well when it fits your performance requirements. There is a lot of code optimization within Veeam for various storage vendor products such as DataDomain.

In case of space savings, in many cases Veeams own optimizations like fast cloning, inline dedup, compression is a poerformance opimized way to nearly achive the same space savings for short term retention.

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 159 guests