Architectural Design of Cloud Applications: a Performance-aware Cost Minimization Approach
ABSTARCT :
Cloud Computing has assumed a relevant role in the ICT, profoundly influencing the life-cycle of modern applications in the manner they are designed, developed, deployed and operated. In this paper, we tackle the problem of supporting the design-time analysis of Cloud applications to identify a cost-optimized strategy for allocating components onto Cloud Virtual Machine infrastructural services, taking performance requirements into account. We present an approach and a tool, SPACE4Cloud, that supports users in modeling the architecture of an application, in defining performance requirements as well as deployment constraints, and then in mapping each architecture component into a corresponding VM service, minimizing total costs. An optimization algorithm supports the mapping and determines the Cloud configuration that minimizes the execution costs of the application over a daily time horizon. The benefits of this approach are demonstrated in the context of an industrial case study. Furthermore, we show that SPACE4Cloud leads to a cost reduction up to 60%, when compared to a first-principle technique based on utilization thresholds, like the ones typically used in practice, and that our solution is able to solve large problem instances within a time frame compatible with a fast-paced design process (less than half an hour in the worst case). Finally, we show that SPACE4Cloud is suitable to model even micro service-based applications and to compute the corresponding optimized deployment configuration which is compared with a state-of-the art meta-heuristic alternative method, achieving savings between 21% and 85%.
EXISTING SYSTEM :
Existing solutions do not fully address all of the pre-ceding challenges. For example, Ernest [37] trains a performance model for machine learning applications with a small number of samples but since its performance model is tightly bound to the particular structure of ma-chine learning jobs, it does not work well for applications such as SQL queries (poor adaptively). Further, Ernest can only select VM sizes within a given instance family, and performance models need to be retrained for each instance family.
DISADVANTAGE :
In addition, trying each cloud configuration multiple times to get around the dynamics in the cloud (due to resource multiplexing and stragglers)would exacerbate the problem even further.
While such data is available to data center operators, it is out of reach for normal users. Cherry Pick works with a restricted amount of data to get around this problem.
PROPOSED SYSTEM :
In this paper, we present Cherry Pick a system that unearths the optimal or near-optimal cloud configurations that minimize cloud usage cost, guarantee application performance and limit the search overhead for recurring big data analytic jobs. Each configuration is rep-resented as the number of VMs, CPU count, CPU speeder core, RAM per core, disk count, disk speed, and net-work capacity of the VM.
ADVANTAGE :
A good cloud configuration can reduce the cost of analytic jobs by a large amount.
The arithmetic mean and maximum running cost of configurations com-pared to the configuration with minimum running cost for four applications across 66 candidate configurations.
|