DescriptionOpenStack has become an ideal solution for managing HPC infrastructure and workloads. However, performance overheads, network integration, and system complexity pose daunting architectural challenges. High performance parallel file systems and the integration of HPC resources into on-premise hybrid clouds allow organizations to increase agility and reduce cost. Using OpenStack technology to control large pools of compute, storage and networking resources throughout the data center allows IT organizations to reduce storage cost while maintaining performance and utilization as it scales. This BoF is aimed at architects, administrators, software engineers, and scientists interested in designing and deploying OpenStack infrastructure for HPC.