Depends on #11
Should move boilerplate part of clustering code (the one which creates workers - with number of workers configurable, AND the one which creates TRIVIALIZABLE error handlers such as "restart worker") into the infrastructure code (configurable via config).
As a part of this issue, we SHOULD move current thread-based clustering here (paving the road for node.js-compatible thread-based fork()).
Depends on #11
Should move boilerplate part of clustering code (the one which creates workers - with number of workers configurable, AND the one which creates TRIVIALIZABLE error handlers such as "restart worker") into the infrastructure code (configurable via config).
As a part of this issue, we SHOULD move current thread-based clustering here (paving the road for node.js-compatible thread-based fork()).