Hi all, Here is a thought on network partition concept that we have also discussed in thread [1].
>From Autoscaler POV, We can consider "network partition" as the "partition group" which we already have in deployment policy. Because this is a same type of grouping that we require in IaaS level. Scale up/ Scale down rules will monitor network partitions separately and act control the load according the deployment policy. Partitions will be selected using partition algorithms defined in Deployment policy. Since the load of one network partition will not affect another, it will be enough to monitor the load independently and act. Minimum check rule will still run against normal partitions and it will make sure that the minimum count is preserved. Please send your thoughts. [1] [Discuss] [Load Balancer] Load Balancer Clusters for Network Partitions Thanks. -- -- Lahiru Sandaruwan Software Engineer, Platform Technologies, WSO2 Inc., http://wso2.com lean.enterprise.middleware email: [email protected] cell: (+94) 773 325 954 blog: http://lahiruwrites.blogspot.com/ twitter: http://twitter.com/lahirus linked-in: http://lk.linkedin.com/pub/lahiru-sandaruwan/16/153/146
