No abstract

Grid middleware and the computational infrastructure that sits beneath it still present a steep learning curve for ordinary scientific end users. Subsequent to the development of grid computing has been the rise of cloud computing, further complicating the process of using third-party computation to execute biomedical simulations. In this paper, we present the Application Hosting Environment 3.0, which virtualizes grid applications and exposes the functionalities through a simple and concise RESTful web service interface. This simple interface allows grid deployed applications to be accessed through simple HTTP calls, meaning that users can easily integrate such applications into more complicated workflow. We also describe how AHE 3.0 is being integrated into a wider biomedical simulation infrastructure, to allow users to seamlessly access cloud and high performance computing resources.

Information technology is revolutionizing medicine. The hospital of the future will rely on advanced ICT infrastructures to aid the treatment of patients, through personalized, predictive simulation. The infrastructure needed to support such simulation techniques comprises systems to store and manage the data, and computational platforms to execute simulations and run data mining algorithms. Many simulation techniques require access to high performance computation, up to and beyond the petascale and will also benefit from the exascale in due course; suitable, usable interfaces are needed to access such resources. In this paper, we present the Application Hosting Environment 3.0, which virtualizes grid applications and exposes the functionalities through a simple and concise RESTful web service interface. We go on to describe how AHE 3.0 is being integrated into a wider biomedical simulation infrastructure, to allow users to seamlessly access cloud and high performance computing resources.