Coalition of Clouds using Grid services and standards

Abstract: Cloud computing is a model for facilitating expedient, on-demand network access to a shared pool of computing resources (e.g., networks, storage, servers, services, and applications) that can be rapidly provisioned and released with minimal management effort or service provider communication. This cloud model endorses availability and is composed of five essential characteristics (On-demand self-service, Resource pooling, Broad network access, Measured Service, Rapid elasticity). Also research communities such as AMO (Atomic, Molecular and Optical) physics research, Biology, and Neuroscience are investigating the relevance of Clouds, with their strengths and weaknesses in scientific environments. In this paper we will illustrate that in scientific environments there are some certain areas where cloud services are supporting the challenging e-Science requirements for virtual communities, resource discovery and dynamic service, resource federation and identity, and access to data catalogues. The Grid technology has actively contributed to address some of these issues, thus we propose to reuse existing Cloud services with Grid technology, experiences and production services, including standardization. In this paper we will provide strategy of how to apprehend the multi-cloud associated deployments based on a survey of existing Grid technologies in context expanding it with lessons learned gained in scientific environments. The contribution spotlight on the areas of data, compute, security and information. We will also show possible benefits that scientists can gain by adopting proposed solutions in cloud-based deployments.

Keywords: AMO; Grid Computing; multi-cloud; Clouds; e-Science