Hi Maksym,

Unfortunately we are currently in the process of deprecating the AIMMS PRO on-premise cluster functionality. The reasons for that are somewhat related to the questions you are asking.

In our cloud environment we have solved these issues because we can simply choose as we see fit. It becomes very complex to come up with a general solution that would work for every environment that people might have on premise. Therefore we decided to discontinue with AIMMS PRO on-premise clusters.

If you are interested in running massive amounts of solve jobs at the same time, you might want to consider doing this in our cloud where you do not need to worry about setting up infrastructure etc. Depending on your business needs we can discuss alternative solutions.

I’m asking because now I involved in migration AIMMS Platform from on premise servers to Azure, and owner want to increase  speed of solving  tasks. I’ve looked at clustered  architecture in this considerations. But as I understand to speed up solving process I just can increase the size of VM amount of resource for platform, am I right? 

Yes, currently with the on-premise (aka customer managed) PRO server the only way to scale capacity is by having bigger instances.