A Joint Benchmark-Analytic Approach for Design-Time Assessment of Multi-Cloud Applications

Verifying that a software system shows certain non-functional properties is a primary concern for cloud applications. Given the heterogeneous technology offer and the pricing models currently available in the cloud market it is extremely complex to find the deployment that fits the application requirements and provides the best Quality of Service (QoS) and cost trade-offs. This task can be very challenging, even infeasible if performed manually, since the number of solutions may become extremely large depending on the number of possible providers and available technology stacks. Furthermore, with the increasing adoption of cloud computing, there is a need for fair evaluation of cloud systems. Today's cloud services differ among others by cost, performance, consistency guarantees, load-balancing, caching, fault tolerance, and SLAs. Moreover, cloud systems are inherently multi-tenant and their performance can vary over time, depending on the congestion level, provider policies, and the competition among running applications. System architects and developers are challenged with this variety of services and trade-offs. Hence, the purpose of a cloud benchmark should be to help developers when choosing the right architecture and services for their applications. In this paper we propose a joint benchmarking and optimization methodology to support the design and migration of legacy applications to Cloud. Our approach is effective in identifying the deployment of minimum costs, which provide also QoS guarantees.