Too BIG to be useful

[Poweruser]

Veeam One is nice, if you have thousands of machines.
But if you have one, two or three its far too big.
But there is a small feature i like to have without installing the BIG one.

The feature i write about is: best practices analyzer for hyper-v hosts.
I want to know what i can improve. Like changing powerplan or sth. else.

Can you include this in a far smaller application?
a small exe file portable or sth. else?
or just give out the list as pdf to check myself?

[Shestakov]

Hello,
Could you specify what looks big for you, .iso file, amount of disk space taken when installed, database size or UI looks massive and hard to get where to go?
There were not such a plans to make a smaller versions since infrastructure size of our average customer is getting bigger, but it's good to get every feedback.
Thanks!

[Poweruser]

Too big means running ONE on 1 host with only 1 physical server.
then its nearly useless.

too big means:
services wasting resources, iis, etc.
disk space
cpu waste
and big menus for nearly no content.

for the essentials license its not that useful..

[Shestakov]

Got it.
However with small infrastructure, Veeam ONE should not consume lots of resources.
If you designed lightweight version of Veeam ONE, what would you have there besides best practices analyzer for hyper-v hosts?
Thanks!

[Poweruser]

I just did a quick test of veam one.

i think its usefull to see: backup grow, backup free disk space for each VM, host (if VHDs are overprovisioned) and repository free space.
also if repository is shared with the VHDs it should be obeyed.

maybe these information should be shown in B&R too.

[Shestakov]

There is an option to choose which metrics you want or don't want to collect. You may also switch to "Backup Data Only" mode.
Thanks for the feedback!

[Poweruser]

Yes, thats nice for medium installations.
But imagine you got one or two hosts and a hand of VMs.
Its like building a big tower for one flat...

[Shestakov]

Most of functionality is useful even if you have one host. Plus the infrastructure may grow and you will need to upgrade from lightweight to full version.
Anyway, your request is clear and taken into account.
Thanks

[Poweruser]

Yes, its nice to see how much space is free, how much space may be used in future, ..
But anyway the tool is far too big, especially if you run it on the main hyper-v host instead in a vm.
this may impact performance.
And yes.. it may grow... but if it grows from 1 to 2 physical host or doubling again to 4 hosts..
its not much better.. and growing two times 2x... is already much

EDIT: thx for taking it into account

[Shestakov]

I've got your point. But based on my experience, if you ask customers with small infrastructures what they would like to have in the lightweight product version, you will get variety of different replies.
Your request is the first in its kind, so let's wait for other opinions.

[Poweruser]

Yes that may vary,
anyway it would be nice to have a list of best practices checkpoints or just a stand alone tool

[tommls]

Put support / best practices information online called "Recommendations for using Veeam ONE in Tiny Environments" where "Tiny" <3 physical hosts and however many VMs are hosted therein, regardless of hypervisor.
What metrics to turn on/off, what reports to use, how to filter/reduce the excessive billions of alerts, etc.
Should not be difficult to write and publish.
Just scale the existing recommendations for sites with billions of host servers down to sites with <= 3 host servers.
Conceptually similar to resizing a recipe for cooking in bulk down to cooking for 1-2 people.

[Poweruser]

yes that could be one solution, but i prefer not to install anything from one except a small portable executable or open an pdf which tells me:
a) dont use pass through drives (snapshot problem)
b) dont user powersave profile on host
c) disable remotedesktop .. (really?)
d) dont overprovision more than 1:1, 1:2
e) disable hyperthreading except for terminalserver only.
f) use faster cores than more cores (really?)
g) assign minimum of x % for primary domain controller vm
h) set start up delays
i) add a shutdown script so hostshutdown will shutdown all vms in the right order and force them after x time (which is not supported by m$ :/!!!)
j) assign at least 2 cores per vm
k) dont assign too many cores, because machine wont run if it cant get all cores at the same time, so it will run less than with less cores.

[Shestakov]

Thanks for the idea, Tom!
How many VMs do you have on those 3 hosts?