A tool to support deployment of scientific software as a service
Resumo
Most of existing e-Science infrastructure is based on shared, interconnected grids. This approach lowers the bar for access to large-scale computational resources and makes possible the collaboration of geographically dispersed teams. However, it requires a big up-front investment and the acquired resources must be continuously maintained and upgraded. Cloud infrastructures are an alternative since resources can be allocated on demand and recently have become more suitable for HPC (High-performance computing). This paper describes our work on the installation and configuration automation of software packages of transient grids based on Infrastructure as Code (IaC).
Referências
Hüttermann, M. (2012). DevOps for Developers, volume 1. Springer.
Kim, H., el Khamra, Y., Jha, S., and Parashar, M. (2009). An autonomic approach to integrated hpc grid and cloud usage. In Proc. Intl. Conf. on e-Science, pages 366–373.
Marschall, M. (2013). Chef Infrastructure Automation Cookbook. Packt Publishing Ltd.
Rehr, J. J., Jorissen, K., Vila, F. D., and Johnson,W. (2012). High-performance computing without commitment: Sc2it: A cloud computing interface that makes computational science available to non-specialists. In Proc. Intl. Conf. on E-Science, pages 1–6.
Singh, G., Kesselman, C., and Deelman, E. (2006). Application-level resource provisioning on the grid. In Proc. Intl. Conf. on e-Science and Grid Computing, pages 83–.
Strijkers, R., Toorop, W., Hoof, A. v., Grosso, P., Belloum, A., Vasuining, D., Laat, C. d., and Meijer, R. (2010). AMOS: Using the cloud for on-demand execution of e-science applications. In Proc. Intl. Conf. on e-Science, pages 331–338.