On Wed, Nov 26, 2014 at 3:19 PM, Akila Ravihansa Perera <[email protected]> wrote:
> Hi, > > Have you considered RIF stats publishing story in a LB clustered scenario? > Each LB instance will be publishing stats in parallel, am I correct to > assume that? > We might need to assume that we got the stats parallel even though there are small time gaps in between. i.e. If we publish events from LB with 10 sec. time interval, we will receive all the events at CEP inside 10 sec. time duration. In that case, will those values get overwritten by each other? > For this, we need to adjust CEP so that it aggregate the values rather than overwrite them. Thanks. > > Thanks. > > On Wed, Nov 26, 2014 at 12:44 PM, Imesh Gunaratne <[email protected]> > wrote: > >> Hi Devs, >> >> This is to discuss the clustering model of the Stratos Load Balancer: >> >> >> >> As shown in the above diagram in the load balancer we need to replicate >> algorithm context and sessions across the members in the cluster. Since the >> load balancer does not persists any contexts we are planning to use carbon >> clustering implementation (which uses Hazelcast) to replicate above data >> structures. >> >> Here the algorithm contexts are managed by the load balancer endpoint >> implementation and sessions are managed by Apache Synapse module. >> >> Thanks >> >> >> >> -- >> Imesh Gunaratne >> >> Technical Lead, WSO2 >> Committer & PMC Member, Apache Stratos >> > > > > -- > Akila Ravihansa Perera > Software Engineer, WSO2 > > Blog: http://ravihansa3000.blogspot.com > -- -- Lahiru Sandaruwan Committer and PMC member, Apache Stratos, Senior Software Engineer, WSO2 Inc., http://wso2.com lean.enterprise.middleware email: [email protected] blog: http://lahiruwrites.blogspot.com/ linked-in: http://lk.linkedin.com/pub/lahiru-sandaruwan/16/153/146
