Hi Valentin, I saw that, the ClusterNode provides access to some interesting metrics but it does not provide the list of the actual active jobs.
I would like to implement load balancing strategy that routes the jobs depending the resources requested by it (cpus, mem, etc) and the actual allocation of the jobs running in each node. Cheers, Paolo On Tue, Feb 23, 2016 at 11:08 PM, vkulichenko <valentin.kuliche...@gmail.com > wrote: > Hi Paolo, > > The task topology is passed directly to LoadBalancingSpi.getBalancedNode() > method: > > public ClusterNode getBalancedNode(ComputeTaskSession ses, > *List<ClusterNode> top*, ComputeJob job) throws IgniteException; > > Is it what you're looking for? > > -Val > > > > -- > View this message in context: > http://apache-ignite-users.70518.x6.nabble.com/Active-jobs-from-LoadBalancingSpi-tp3157p3160.html > Sent from the Apache Ignite Users mailing list archive at Nabble.com. >
