Last Friday Veeam released version 8 of their Backup and Replication product and like the previous releases the features keep on coming with improvements to an already rock solid backup platform continuing.


Most exciting from my point of view is the debut of Veeam Cloud Connect which I have been lucky enough to be beta testing since June of this year. As a Veeam 8 Cloud Connect Launch partner i’ve had my hands on two betas and an RC of the new version to get a good early look at. And while I focused on Cloud Connect below are a couple of the standout features of v8 i’d like to highlight.

Backup I/O Control:

Ensures workload’s availability by reducing the impact of backup and replication
jobs on production VMs running on the same storage where backed-up VMs reside by monitoring the production datastore read latency and controlling job I/O to keep the latency acceptable limits. Backup I/O Control works in conjunction with Automatic Load Balancing to ensure no new intensive tasks such as virtual disk backup, restore or snapshot removal are assigned to the same datastore if the datastore’s latency is above a user-defined first latency threshold.


To ensure that backups can complete under any circumstances, Backup I/O Control will always allow at least one active task per datastore, and will never throttle any given task to less than 10% of full I/O capacity.

As you can see above you need Enterprise Edition to use this feature.

Forever forward Incremental backup mode:

This is a new default backup mode creates forward incremental backups for all the following runs after the initial full backup, which always remains the oldest restore point in a full backup chain. When the retention policy needs to remove the oldest restore point, the job merges the oldest incremental backup into the full backup file, discarding any data replaced in the full backup file by this process.

Because forward incremental backup creation involves mostly sequential writes, the forever forward incremental backup mode reduces the time a VM runs off of a snapshot by up to 3x when compared to the reverse incremental backup mode. This prevents a VM snapshot from growing large and making its commit much faster reducing both backup window and load on production storage.

Job start time priority:

The backup infrastructure resource scheduler will now prioritize all jobs according to their start times, and attempt to finish any particular job as soon as possible once it has been started. This means that if you start multiple jobs at once, the scheduler will assign newly appearing processing resources to the job that started earlier, as opposed to assigning them between all running jobs. This approach ensures that the backed-up state of all VMs within the same job remains as close together as possible

Snapshot Hunter:

This detects and automatically consolidates hidden VM snapshots to prevent production VMs from stopping due to datastores filling up. To detect hidden snapshots before exiting the jobs will physically scan datastores for snapshot files belonging to snapshots not registered in vSphere.  If hidden snapshots are found, the job will attempt to automatically consolidate them. If the job fails to remove hidden snapshots due to locking, the job will log an event and exit, however a background system process will attempt to perform consolidation three more times every four hours in the hope that file locks will be removed. If these attempts fail Snapshot Hunter will stop trying, and a warning email will be sent to the global notification recipients urging them to take manual action.

Hot add improvements:

Backup proxies can now process multiple virtual disks of the same VM in parallel using hot add. This includes hot add backup, restore and replication (on both source and target proxy). Additionally, performance of operations can see an increase of up to a few times depending on the environment.

CBT Bug Fix:

As explained in this post from Luca Dell’Oca the bug that was discovered a few months back by VMware has been addressed by B&R being able to automatically reset CBT informations on a processed VM upon detecting a virtual disk configuration change.

Job Retry Suppression:

The last new feature is a small one, but one that’s handy for Service Providers. Previously if a job failed an email would be generated upon every failure and subsequent retry which tends to make customers edgy upon seeing backup failures.



As seen above there is a new checkbox in the SMTP Settings that lets you suppress notifications until the last retry…a nice touch!