Conference paper Open Access
Antonenko Vitaly; Petrov Ivan; Smeliansky Ruslan; Huang Zhen Chun; Chen Min; Cao Donggang; Chen Xiangqun
Recently in many scientific disciplines, e.g. physics, chemistry, biology and multidisciplinary research have shifted to the computational modeling. The main instrument for such numerical experiments has been supercomputing. However the number of supercomputers and their performance grows significantly slower than the growth of users’ demands. As a result, users may wait for weeks until their job will be done. At the same time the computational power of cloud computing grow up considerably and represent today a plenty available resources for numerical experiments for many applications. There are several problems related to cloud and supercomputer integration. First, is how to make a decision where to send a computational task: to a supercomputer or to cloud. Second, various platforms may have significantly different APIs, and it’s often labor-expensively for researchers to move from one platform to another since it would require large code modification. In this research we present MC2E –an environment for academic multidisciplinary research. MC2E aggregates heterogeneous resources such as private/public clouds, HPC clusters and supercomputers under a unified easy-to-use interface. This environment will allow to schedule parallel applications between clouds and supercomputers based on their performance and resource usage. MC2E will also provide a Channel-on-Demand service to connect clouds and supercomputers with channels that are used to send data for parallel applications.