Abstract: Disclosed is a method that includes collecting first temporal statistics for a port element in a computing environment, collecting second temporal statistics for a switch element in the computing environment, collecting third temporal statistics for the computing environment generally, computing a spatial correlation between network features and network elements comprising the port element and the switch element and computing, via a machine learning technique, a port dynamic weight for the port element and a switch dynamic weight for the switch element. The method can also include scheduling workload to consume compute resources within the compute environment based at least in part on the port dynamic weight for the port element and the switch dynamic weight for the switch element.

Abstract: Disclosed is a method that includes collecting first temporal statistics for a port element in a computing environment, collecting second temporal statistics for a switch element in the computing environment, collecting third temporal statistics for the computing environment generally, computing a spatial correlation between network features and network elements comprising the port element and the switch element and computing, via a machine learning technique, a port dynamic weight for the port element and a switch dynamic weight for the switch element. The method can also include scheduling workload to consume compute resources within the compute environment based at least in part on the port dynamic weight for the port element and the switch dynamic weight for the switch element.

Abstract: Disclosed is a method that includes collecting first temporal statistics for a port element in a computing environment, collecting second temporal statistics for a switch element in the computing environment, collecting third temporal statistics for the computing environment generally, computing a spatial correlation between network features and network elements comprising the port element and the switch element and computing, via a machine learning technique, a port dynamic weight for the port element and a switch dynamic weight for the switch element. The method can also include scheduling workload to consume compute resources within the compute environment based at least in part on the port dynamic weight for the port element and the switch dynamic weight for the switch element.

Abstract: A network controller deploys network management applications to containers hosted on compute nodes. For each container on each compute node, the controller determines the network connectivity required by the application; assigns to the container one or more Internet Protocol (IP) addresses that satisfy the network connectivity, and provides the one or more IP addresses to the container; upon detecting the container is lost, blocks the one or more IP addresses so that the they cannot be re-assigned to any other container; and upon detecting that the application has been re-deployed to a new container on a new compute node, unblock the blocked one or more IP addresses, and assign and send the one or more IP addresses to the new container to enable the new container to establish the network connectivity required by the application without network address translation.

Abstract: Disclosed is a method that includes collecting first temporal statistics for a port element in a computing environment, collecting second temporal statistics for a switch element in the computing environment, collecting third temporal statistics for the computing environment generally, computing a spatial correlation between network features and network elements comprising the port element and the switch element and computing, via a machine learning technique, a port dynamic weight for the port element and a switch dynamic weight for the switch element. The method can also include scheduling workload to consume compute resources within the compute environment based at least in part on the port dynamic weight for the port element and the switch dynamic weight for the switch element.

Abstract: A network controller deploys network management applications to containers hosted on compute nodes. For each container on each compute node, the controller determines the network connectivity required by the application; assigns to the container one or more Internet Protocol (IP) addresses that satisfy the network connectivity, and provides the one or more IP addresses to the container; upon detecting the container is lost, blocks the one or more IP addresses so that the they cannot be re-assigned to any other container; and upon detecting that the application has been re-deployed to a new container on a new compute node, unblock the blocked one or more IP addresses, and assign and send the one or more IP addresses to the new container to enable the new container to establish the network connectivity required by the application without network address translation.