Workload Trace Generation for Dynamic Environments in Cloud Computing

Cloud computing datacenters provide millions of virtual machines in actual cloud markets. In this context, Virtual Machine Placement (VMP) is one of the most challenging problems in cloud infrastructure management, considering the large number of possible optimization criteria and different formulations that could be studied. In the existing literature, the VMP problem is mostly formulated as a provider-oriented VMP considering static (offline) or dynamic (online) formulations. Considering the on-demand model of cloud computing, the VMP problem should be optimized dynamically to efficiently attend typical workload of modern applications. This work proposes a novel taxonomy in order to understand possible challenges for Cloud Service Providers (CSPs) in dynamic environments, based on the most relevant dynamic parameters studied so far in the VMP literature. The identified challenges must be addressed in future formulations and novel algorithms for an efficient utilization of resources in real world cloud computing datacenters and for a further advance of this important research area. I. PROPOSED TAXONOMY A. Cloud Service Notation Formally, a cloud service Sb can be distributed across different possible cloud datacenters. Each cloud datacenter DCc hosts VMs V ′ cj associated to different cloud services. A VM V ′ cj associated to Sb is denoted as V ′′ bcj . where: Sb: Cloud service b; DCc: Cloud datacenter c; mDCc: Number of VMs Vj in DCc; mSb: Number of VMs Vj in Sb; V ′ cj : Vj in DCc; V ′′ bcj : Vj in DCc from service Sb. B. Dynamic Environment Notation In this work, different environments represent particular types of requests for placement of cloud services from the CSPs’ perspective. According to the proposed taxonomy, 16 different possible dynamic environments could be identified considering one or more of the following classification criteria: (1) horizontal elasticity, (2) vertical elasticity, (3) overbooking of server resources and (4) overbooking of network resources. Additionally to the notation presented in Section I-A, each cloud datacenter DCc is defined as: DCc = {V ′ c1, V ′ c2, . . . , V ′ cmc } (1)