As usual with any new release, Anton Gostev, Director of Product Management at Veeam (or better, the man behind Veeam Backup & Replication), has just published the complete “What’s new” paper about the upcoming latest version of their product.
If you can’t wait to read it, you can download it here.
As always, marketing has placed the focus on the new biggest features since months, but there are also several “gems” more or less hidden, that are now finally announced in this document. Has I already did in the past, I would like to highlight the ones I like the most, with a small comment. In this last release the number of enhancements is insane!
– Parallel processing of multiple VMs = now Veeam can process several VM inside a single job at the same time. Before, a single job could have been made with several VMs, but each of them was processed after the previous, one by one. There has been situations (I had some customers with this problem) where a user wanted to save a bunch of VMs all at the same time: the only solution was to create several jobs, and start all of them at the same time. Often, there were already other jobs running filling the proxies or limits set by cuncurrenct configuration, so the different jobs ended up by not beeing processed at the same time. now this new option offer a new and improved way to process VMs in this way.
– Parallel processing of virtual disks within VMs = this is one of the coolest “gem” in my opinion. If you have sized accordingly your “data pipe” (source, proxy and target) you can now save all the disks of a single VMs at the same time. This brings a reduction in backup times, and most of all reduces the time a snapshot of a VM is opened. If you think about huge VMs with medium or high I/O activity, this allows to limit the problems around snapshots commit at the end of backups!
– 64-bit data movers = Can I say it? Finally! Before Veeam 7 data movers were 32 bit binaries, and this could lead to problems in RAM management for example, especially in order to manage large backup files (I talked about it in this post). With the new 64 bit data movers those limits are now raised, but be aware you would need to update all the components to Windows 2012 to get this advantage.
– Enhanced I/O pattern for data writes = Veeam has lowered the fragmentation incurring inside the backup file. In this way backup and restore operations’ performances are increased, and you can use with even more peace of mind the Reverse Incremental mode, since it always “reuses” the blocks inside the same backup file.
– Transparent backup of vCenter Server VM = OLE!!! In the past, you had to use some workarounds to backup a virtualized vCenter. Now you can manage it just like any other VM.
– Changed Block Tracking. Changed block tracking can now be enabled for VMs that already have snapshots = this enhancement speaks for itself. I do not know if this has been possible thanks to some new feature from VMware leveraged by Veeam, or was directly created by Veeam. What I do know is that it was not possible to enable CBT on a VM with an active snapshot. Regardless how they manage to solve this problem, thanks!
– Enumerated infrastructure objects = if you ever used Veeam in a large vSphere environment, you already know what’s the issue. Veeam enumerates the objects it has to save at the beginning of a job; this operation can take quite some time on large environments; multiply this time for all the jobs created, and you are loosing a valuable amount of time. Now you can save that time with this new feature.
– File-level recovery UI is no longer modal = now you can iconize the recovery UI and do other activities, without beeing forced to wait for the end of a recovery task
– NSS (Novell Storage Services) = I followed the birth of this feature on Veeam forums, and is again another proof of how this company is totally focused on their customers. It all started with some users requesting NSS support, then Veeam released some beta versions to be tested, and now NSS support is native into the main product. Kudos to Veeam!
– Credentials Manager = I have a post coming out when the NDA will finish right about this feature. Basically, now you can save all the credentials in a dedicated and centralized area of the software, and when you update them, each part of the software using it will have the updated credentials without having to edit each job.
– Storage Integration with Plug-in architecture = has you all know since many months, actually Veeam supports only HP 3Par and HP StoreVirtual. But the plug-in architecture paves the way for the support of new and upcoming supported storage.