Paolo, I think you can create a replicated cache and store such metrics there. When a job starts, it increments number of used CPUs and decrements when it finishes execution. Load balancer can make decisions based on data in the cache.
Will this work? -Val -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Active-jobs-from-LoadBalancingSpi-tp3157p3175.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.