A blocking probability estimator for the multi-application and multi-resource constraint problem

The provisioning of network equipment is becoming increasingly challenging due to the large number of application types that must be supported at once. In addition, most applications must make use of various types of resources, including network bandwidth and server CPU cycles, among others. An application request may be blocked when one or more of the resource types that are required to achieve the desired QoS cannot be reserved due to their shortage. This problem is referred to as the multi-application and multi-resource (MA-MR) constraint problem. In this paper, a Markov chain model is proposed to efficiently and accurately estimate the blocking caused by the MA-MR constraint. The model´s strength is its scalability and ability to account for hundreds of application types concurrently sharing multiple pools of distinct resources. The proposed blocking probability estimator is applicable to a number of practical engineering tasks. For instance, in the cloud infrastructure, the estimator may enable rapid decisions to be made in real-time while accounting for the blocking probability that application requests may experience due to a multitude of concurrent resource constraints, including the lack of network bandwidth, server CPU cycles, memory and storage.