I have a question. I know that we have Ignition.setClientMode(...) parameter which controls whether local node will participate in caching as a data node, or will simply access the data as a client from remote nodes.
Do we have the same support for the compute grid. For example, do client nodes participate in job execution by default? Or does user need to create ClusterGroup.forClients() cluster group and pass it into compute? I am OK either way, but I need to document it properly. Alexey G, since you added the "client" node feature, can you advise? D.
