With the medicine technology advances, it is possible to create a patient profile by a full genomic data along with vast amount of patient data from personal fitness devices, medical record, etc. Translational medicine is aim at moving faster from research to patient care by integrating the data across traditional silos which requires a cloud to support multi-tenancy, self-service, big data analysis and HPC workload. OpenStack are good at virtual machine management, while we use VM to serve as a container host which brings virtualization overhead in overlay networking, I/O operations, etc. and it won’t get benefit from some advanced features. In this session, we'll share:
- How to build an OpenStack cloud for HPC with multi-tenancy and self-services;
- How to build a bare metal kubernetes cluster with OpenStack Magnum and Ironic;
- The issues we are facing and how we overcome them;
- The gaps remaining and what does the community need to improve in the future;
- More thinking and new ideas.
- How does OpenStack resolve the issues of HPC in Translational Medicine Research.
- How to do a 1-key bare metal container cluster provision on OpenStack with Magnum and Ironic projects.
- Some new thinking and improvements with current container management on OpenStack.