Grid and Cloud Computing Methods in Biomedical Research

Cloud computing is currently on everyone’s lips – it is of interest to researchers (to speed up computations), managers (to reduce costs), and also the general public (to access music everywhere). The information technology research and advisory firm Gartner Inc., for example, announced that cloud computing will be one out of ten top information technology topics in the next years. At least in 2015 cloud computing is predicted to be a mass product and will likely be the preferred solution for many application areas [1]. In view of these developments, the idea of cloud computing sounds promising for the communities of Medical Biometry and Medical Informatics as well. Possible benefits include cost reductions in hardware, maintenance and software licensing, access to latest software and hardware for everyone from any place in the world, standardized and centralized analyses platforms, and new ways of collaboration between researchers and practitioners. Biomedical research, as many other scientific disciplines, currently experiences the increased focus on high performance computing (HPC) – mainly driven by the three aspects of 1) larger data sets and the “big data” hype, 2) increased computational requirements stemming from more sophisticated statistical methodologies, and 3) new computing environments including multi-core processors and the use of graphics processing units to do general purpose computing. On this account, the German working group for Statistical Computing of the German Region of the International Biometric Society (IBS-DR) and the German Association of Medical Informatics, Biometry and Epidemiology (GMDS) organized two workshops in the year 2011 (March 23 in Munich, and September 27 in Mainz). In the course of these meetings, whose organizing committees consisted of the authors of this editorial, leading community experts discussed how HPC technologies can be used to improve and accelerate biomedical analysis. The present Focus Theme summarizes the results and the outcomes of the workshops. Specifically, the four papers highlight a selection of benefits and challenges for grid and cloud computing, and the general use of graphic processors in the communities of Medical Biometry and Medical Informatics. Additionally, they contain evaluation of technical solutions and business model aspects. Before going into details about the four papers, we briefly introduce common terms and relations of HPC technologies.