Introduction

Recent developments in genome sequencing and post-genomic technology have resulted in an enormous step change in both the rate of whole genome sequence deposition in the databases and the rate at which large scale biological datasets are being generated. The flexible analysis of these huge datasets raises problems for the ordinary bioinformatician who does not have access to their own massive, dedicated compute clusters. The development of Grid technology promised to allow access to remote processing power for computationally demanding applications in bioinformatics. This project aims to fulfil this potential by providing a development system for the bioinformatician (who will in turn use it to support the biologist).  The resulting architecture harnesses Grid technology in a scalable fashion, and also eases the process of its use, devolving the need for the developer to worry the details of job deployment, such as task scheduling.

A fundamental consideration in the development of Microbase v1.0 is to design and implement a set of components that potentially can exist anywhere within a distributed computing environment as one or more instances, and that can, wherever possible, be used in isolation or in combination with another Grid system.

Inter-component communication between is mediated via Web-service interfaces. Scheduling, component orchestration. Some aspects of communication are achieved by the Microbase Grid notification system which also operates via Web service interfaces. Grid technology, by its nature may be difficult to configure, use and deploy. We are trying to ensure that ease of installation and use is one of the primary characteristics for the Microbase system.

The Microbase v1.0 architecture is described more fully in our documentation.