International Journal of Cloud-Computing and Super-Computing
Volume 4, No. 1, 2017, pp 9-14 | ||
Abstract |
Cloud Computing to Develop Applications
|
Commercial clouds bring an outstanding chance to the scientific computing space. Scientific applications typically would love important resources, but not all scientists have access to spare high-end computing systems. Cloud computing has gained the eye of scientists as a competitive resource to run HPC applications at a possibly lower value. However as a completely unique infrastructure, it's unclear whether or not or not or not clouds ar capable of running scientific applications with an affordable performance per cash spent. This work provides a comprehensive analysis of EC2 cloud in varied aspects. we have a tendency to tend to tend to initial analyze the potentials of the cloud by evaluating the raw performance of assorted services of AWS like cypher, memory, network and I/O. supported the findings on the raw performance, we have a tendency to tend to tend to then well worth the performance of the scientific applications running at intervals the cloud. Finally, we have a tendency to tend to tend to match the performance of AWS with a private cloud, soon find the root reason behind its limitations while running scientific applications. This paper aims to assess the power of the cloud to perform well, what is more on price the definitely worth the worth of the cloud in terms of each raw performance and scientific applications performance. Moreover, we have a tendency to tend to tend to value other services alongside S3, purpose and Dynamo DB among several AWS services so on assess the skills of these to use by scientific applications and frameworks. We have a tendency to tend to tend to additionally evaluate real scientific computing application through the Swift parallel scripting system at scale. Armed with each elaborated benchmarks to measure expected performance associated associate complete cash analysis, we have a tendency to tend to tend to expect this paper unit a direction book of facts for scientists to assist them decide wherever to deploy and run their scientific applications between public clouds, personal clouds, or hybrid clouds.