Algorithm and differentiated weights for resource constrained processes

Abstract

The techniques disclosed herein enhance the functionality of network computing infrastructure in resource constrained processes. This is accomplished by assigning differentiated weights to instances of a software service based on the role of the instance. In the context of the present disclosure, a role is a defined set of functionalities within a software service. An individual weight quantitatively represents the computing resource demand imposed by the functionalities of the role. A software orchestration system subsequently places the instances of the software service within a computing environment (e.g., a node, a cluster) for execution. As such, the computing environment can include a resource constraint that represents the capacity of the constituent computing resources to execute the instances of the software service. Accordingly, the instances are placed such that the sum of the weights of the instances is less than or equal to the resource constraint.