scala - Worker dial-in pattern with multiple masters? -


there's worker dial-in pattern described akka, particularly here: http://letitcrash.com/post/29044669086/balancing-workload-across-nodes-with-akka-2. describes way spread load between multiple remote workers. assumes there's 1 master, , workers discover , register it. there way support multiple masters worker dial-in pattern, supports fair , deterministic sharing of workers between multiple masters?

i imagine following situation. let's there's cluster 2 different node roles: front-end , worker. there multiple front-end nodes run http servers. front-ends delegate business logic actors running on worker nodes. front-ends behind simple http round-robin load balancer (nginx).

i'd have shared pool of worker nodes can used of front-ends. if 1 node has more load other, should consume more worker nodes' capacity. if load heavy, should able add more worker nodes (probably automatically via auto-scaling), , should, again, support of front-ends fairly, on need basis.

there couple of naive implementation leading different deficiencies. if workers somehow decide single front-end support, worker capacity might not spread fairly, because front-end load highly dynamic. alternatively, if workers register of front-ends, there might race condition when multiple front-ends request work single worker. in all, don't see way of supporting this. has better idea?


Comments

Popular posts from this blog

Command prompt result in label. Python 2.7 -

javascript - How do I use URL parameters to change link href on page? -

amazon web services - AWS Route53 Trying To Get Site To Resolve To www -