HPC OpenStack Cloud with Optimized Software-Defined Storage

High Performance Computing (HPC) clouds offer researchers access to powerful computing resources with the ability to handle a wide range of workloads without compromising the performance. As these workloads are diverse in nature, the traditional Lustre parallel file system does not suit mostly for a scalable storage solution. Ceph, the software defined storage is a distributed storage system that is designed to provide scalable, highly available storage for a variety of use cases, including HPC. One of the challenges with using Ceph in HPC environments is that HPC workloads often have very high performance and bandwidth requirements, and Ceph's performance can depend heavily on the underlying hardware and network infrastructure. We propose an optimization method for HPC Cloud ecosystem with OpenStack and optimized storage deployment using Ceph software-defined storage as the best option for their needs and to offer an efficient persistent storage for both VMs and container workloads.