An Elastic Middleware Platform for Concurrent and Distributed Cloud and Map-Reduce Simulation-as-a-Service

Middleware platforms empower the developers and users with an additional layer, enhancing scalability and performance. Cloud computing offers variety of resources and involves a tremendous amount of entities. Due to the limited access, along with varying availability of cloud resources, prototypes are often tested against cloud simulation environments. While computing has been evolving with multicore programming, MapReduce paradigms, and middleware platforms, cloud and MapReduce simulations still fail to exploit these developments themselves. This paper describes the research for the de-sign, development and evaluation of a complete fully paral-lel and distributed cloud simulation platform (Cloud2Sim), which tries to fill the gap between current simulations and the actual technology that they are trying to simulate. First, Cloud2Sim provides a concurrent, distributed and elastic cloud simulator, by extending CloudSim cloud simula-tor, using Hazelcast in-memory key-value store. Then, it also provides an assessment of the MapReduce implementations of Hazelcast and Infinispan, as well as elastically distribut-ing their execution in a cluster, providing means of simulating MapReduce executions. A dynamic scaler solution scales out the cloud and MapReduce simulations to multiple nodes run-ning Hazelcast and Infinispan, based on load. The distributed execution model and adaptive scaling solution could further be leveraged as a general purpose auto-scaler middleware for a multi-tenant deployment. ı̈ż£ 1