The HPS3 Service: Reduction of Cost and Transfer Time for Storing Data on Clouds

This is a post-peer-review, pre-copyedit version. The final authenticated version is available online at: http://dx.doi.org/10.1109/HPCC.2014.38[Abstract] In the past several years, organizations have been changing their storage methods as the volume of data they managed has increased. The cloud computing paradigm offers new ways of storing data based on scalability and on good conditions of reliability and accessibility. This paper proposes the design of HPS3, a service that uses compression and concurrency techniques in order to reduce storage costs and data processing times in public storage providers. Different strategies for compressing and uploading data depending on differential characteristics of the datasets are also explained. The evaluation of HPS3 shows, in comparison with the default use of the cloud storage provider (Amazon S3), an average reduction of 61.6% in data transfer volume, 85.5% in upload time and 73.2% in incurred costs. Compared to Drop box, HPS3 shows an average improvement of 27.4% in data transfer volume and 93.6% in upload time.Ministerio de Economía y Competitividad; TIN2013-42148-PXunta de Galicia; GRC2013/05