Have you even wondered how it felt to foresee the future like Cassandra?
Well, last week I felt a little bit like her… You are going to find in the next few days many articles about the new features of Veeam Backup & Replication 7, kept secret until today. I would like to tell you my own side of the story…
First of all, what are we talking about? Veeam had kept two hidden and secret features to be announced last, after the 7 they already leaked in the last months.
One of them was something totally expected. After introducing in the 6.5 version the Veeam Explorer for San Snapshots (VESS), allowing restores from SAN Snapshots of a LeftHand storage, in this new version VESS adds support for 3Par storage, already announced, but most of all the possibility to do backups of the VMs already saved inside those snapshots. Lower impact on running VMs, VMware snapshots not involved at all, a huge benefit in the final RPO.
Sounds interesting? I talked about it last 29th January, in the post It’s time for “Storage Aware” backups?. This is how it works in Veeam:
Read my post, read Veeam release notes, and look how near I was in my prediction!!
If this is probably an easy one, the second is for sure even better!
The second hidden feature is the WAN Accelerator, but for my long time readers, you can call it VeeaMover, as I did in my post dated 22nd January 2013. Here the details of my prevision has been almost embarassing: introduced as the 4th role in the Veeam Backup architecture, the WAN accelerator will allow to copy chosen VM backups (so not forced to use the same backup sets in the primary site) in other storage, with totally different frequencies and policies than those in the primasy site, all in a forever incremental format.
It has been called WAN accelerator because Veeam decided to develop internally the caching and replica features, and not to use an existing solution. This is how it works:
to manage this new feature, you will need as I said (and predicted!!!) to enable a fourth role in the Backup Infrastructure:
Another interesting element, is the new GFS rotation schema available in the WAN Accelerator, so the remote repository can have different copies of the VMs, less and less frequent as time goes by:
For sure, the VeeaMover… ops I meant the WAN Accelerator is an enhancement which was heavily missed. Veeam has remedied this with a tool that seems very well implemented. Veeam aleady has a map of blocks of each VM in each backup, to make a forever incremental copy of these same blocks was definitely possible for them.
Looking at my comments earlier this year, it seems the hub-and-spoke is possible too, given that by distributing different WAN Accelerators in different locations and activating different copy jobs I guess you can create several copies at the same time. In addition, a GFS scheme definitely allows you to optimize the disk space of a remote storage, since unlike the primary backup where you need to keep all of your most recent backups, you rather keep older versions of your backup to meet different retention policies.