Not long ago, High Performance Computing technology was really only viable in highly specialized research labs and universities running special projects. Picture mysterious room-sized machines and technicians in white lab coats…
But today, use of HPC has spread rapidly across all sectors and addresses a wide variety of computing challenges. The once far-fetched goal of exascale computing is no longer that far away and the power of HPC is well understood by IT professionals across the industry. HPC now touches all of us everyday, from preparing for the weather to the designing and fueling of our cars to the guidance we get about our financial investments and to development of the medicines we may rely on every day.
But we’re still sitting far short of the true potential for HPC to change the industry and our lives. Right now, we are at a key inflection point and to reach the next level we now need HPC software capabilities that can take full advantage of the most powerful hardware we produce today but are also based on standards that allow rapid adoption and usage. We need an intelligent HPC software stack that surfaces new advances in I/O processing and cluster & system management but can be re-used for numerous purposes. If we think of the super-computing hardware as the physical brain, this new HPC software is the spark that brings it to life and helps us utilize a much larger percentage of that brain than ever before. This in turn ignites even greater innovation in a rapid and iterative advance towards the future.
OpenHPC seeks to ignite that spark and take HPC to the next level through the power of open source collaboration. By creating a opensource software stack that is both powerful enough for the next generation of HPC and standardized enough for rapid use and adoption, vendors and researchers can focus on applying the power of HPC rather than constantly duplicating the same efforts to build HPC infrastructure. For anyone involved in HPC, you already know how resource and time intensive these efforts can be – you can spend months or even years just building infrastructure that is likely outdated by the time you’re ready to move to production use. With the OpenHPC stack, you can leap forward with a validated, tested and secure infrastructure software stack and get to product in a fraction of the time. This new approach will drive rapid advances in the state of the art and accelerate our progress towards true exascale computing.
SUSE, as a founder and board member of OpenHPC, contributes not just to the Linux OS elements of the project but also to the underlying HPC system building componentry and system tools that can bridge today’s infrastructure development gaps and help move vendors and researchers rapidly towards powerful HPC solution stacks.