Remaining capacity for each node.
Remaining capacity for each node. The value is between 0.0 and 1.0, where 0.0 means no remaining capacity (full utilization) and 1.0 means full remaining capacity (zero utilization).
How much extra weight to give to the stolen time.
Java API: get the singleton instance
The weights per address, based on the capacity produced by the nodeMetrics.
The weights per address, based on the capacity produced by the nodeMetrics.
Converts the capacity values to weights.
Converts the capacity values to weights. The node with lowest capacity gets weight 1 (lowest usable capacity is 1%) and other nodes gets weights proportional to their capacity compared to the node with lowest capacity.
MetricsSelector that uses the combined CPU time metrics and stolen CPU time metrics. In modern Linux kernels: CpuCombined + CpuStolen + CpuIdle = 1.0 or 100%. Combined CPU is sum of User + Sys + Nice + Wait times, as percentage. Stolen CPU is the amount of CPU taken away from this virtual machine by the hypervisor, as percentage.
Low CPU capacity => small node weight.