Parallelization with multiple Academic licenses

[DSarkar]

Hi,

I was wondering if it is possible to parallelize a job on multiple nodes with different Academic licenses installed on each node. If so, how may it be done? Also, my cluster nodes each have 2 processors with 8 cores each. While running the job, if I change the machine settings to Multiprocess, choose No. of processors=2, Enable threading, enable auto-threading, it would take longer than if I run on a single core.  How can I check the actual resources that I used? I appreciate your replies very much.

Regards,

Debarghya

[Jess Wellendorff]

No, this is not possible. The slave nodes must have have same license configuration as the master node does.

[Anders Blom]

Having such a license configuration would be highly unusual... If you have a trial license, it's valid on any machine (so, it's not "different" on the different nodes).

[DSarkar]

Okay thanks for the clarification.