Service Oriented Cloud Storage Performance Analysis

Cloud for big data analytics is an end-to-end solution covering hardware architectures, software platforms, virtual machines, networks, securities and the design of services. Through centralized management and shared computing and storage resources, a cloud platform helps customers to solve the problems of traditional clusters and allows them to improve O&M efficiency and create a truly mobile office. Cloud servers are composed of cloud OS and virtualized platforms. The computing devices usually include multi-core CPU, GPU, FPGA and other multiprocessing facilities. Smart storage engines, intelligent networks, SSD caching mechanisms and other innovations work together to achieve high performance. Cloud based big data analytic platform is not a one-size-fits-all solution. Businesses with varying needs and budgets determine the strategies to create the services in cloud environments. In this talk, we introduce a service oriented storage performance model (service model). Based on the service model, data localities need to be designed to analyze the data either in a cloud data center or in edge systems and client devices. By focusing on handling the critical design constraints at each level of a cloud, optimized services can be approached based on the service model in order to achieve the best performance. The functionality and capability of our methodology on the service model have been validated through benchmarks and measurements on real cloud.

Learning Objectives

Big data service model on cloud
Cloud storage
Best practices architecture for big data computing on cloud
Performance analysis