CYCLONE: A Platform for Data Intensive Scientific Applications in Heterogeneous Multi-cloud/Multi-provider Environment

摘要

This paper presents results of the ongoing development of the CYCLONE as a platform for scientific applications in heterogeneous multi-cloud/multi-provider environment. The paper explains the general use case that provides a general motivation for the CYCLONE architecture and provides detailed analysis of the bioinformatics use cases that define specific requirements to the CYCLONE infrastructure components. Special attention is given to the federated access control and security infrastructure that must provide consistent security and data protection for distributed bioinformatics data processing infrastructure and distributed cross-organisations collaborating teams of scientists. The paper provides information about selected use cases implementation using SlipStream cloud automation and management platform with application recipe example. The paper also addresses requirements for providing dedicated intercloud network infrastructure which is currently not addressed by cloud providers (both public and scientific/community).