For option 1, you could use the advanced multi-host virtual lab and ensure DRS is enabled on the cluster as this would at least introduce some randomness.
For option 2, while originally Veeam did recommend a proxy per host for VMware VSAN, this is not the recommendation at this time. Rather, just calculate how many proxy tasks you need as per the best practices guide and then ensure you scale as needed.
SureBackup
What you are looking for is not implemented in GUI, but it is easily possible by running a script. There are scripts out there you can use to create your own setup. For a example: a customer runs >1500 VMs and runs SureBackup 4 times a day. Each run a bunch of new VMs are selected.
vSAN
To run a proxy on each node reduces network-load, because Veeam prioritize the proxy that is located on the same node (data locality) as the VM to backup. But this is not a must. Veeam can also backup VMs located on different nodes than the proxy. When you operate more than two nodes, even hosts can run VMs without having data of these VMs local.
https://www.veeam.com/blog/deep-dive-on-vmware-vsan-6-6.html
Point 1. Is possible, but you need the relevant VMware licensing (assuming we’re talking VMware here as you mentioned vSan in point 2) for Distributed Switching as a pre-requisite for Advanced Multi-Host.
There’s already other good answers here for point 2.
For option 1, you could use the advanced multi-host virtual lab and ensure DRS is enabled on the cluster as this would at least introduce some randomness.
For option 2, while originally Veeam did recommend a proxy per host for VMware VSAN, this is not the recommendation at this time. Rather, just calculate how many proxy tasks you need as per the best practices guide and then ensure you scale as needed.
Need to know more about option 2 solution you mentioned.
For option 1, you could use the advanced multi-host virtual lab and ensure DRS is enabled on the cluster as this would at least introduce some randomness.
For option 2, while originally Veeam did recommend a proxy per host for VMware VSAN, this is not the recommendation at this time. Rather, just calculate how many proxy tasks you need as per the best practices guide and then ensure you scale as needed.
Need to know more about option 2 solution you mentioned.
In my Topic, I explained all you need for know how much resource need to the proxies.
https://community.veeam.com/blogs-and-podcasts-57/veeam-architects-site-103?postid=156#post156
For option 2
To transport data of VMs residing on VSAN in the Virtual appliance mode, you must assign the backup proxy role to a VM.
The backup proxy VM must meet the following requirements:
The backup proxy VM must reside on any of ESXi hosts connected to a VSAN cluster.
Veeam Backup & Replication will retrieve data of processed VMs over the I/O stack of the ESXi host on which the backup proxy is deployed.
Disks of the backup proxy VM must reside on the VSAN datastore.
If you have several backup proxies on ESXi hosts in the VSAN cluster, Veeam Backup & Replication chooses the most appropriate backup proxy to reduce the backup traffic on the VSAN cluster network. To choose a backup proxy, Veeam Backup & Replication checks HDDs directly attached to every ESXi host and calculates the amount of VM data on these HDDs. The preference is given to the ESXi host that has a direct access to an HDD with the maximum amount of VM data. This approach helps reduce workload on the ESXi I/O stack during data transport.
@eprieto This is great. Thanks
