Achieving Optimal Price/Performance for Your HPC Workload
A guide to discovering the best AWS instances and configurations for your use case.
January 2, 2020
High Performance Computing (HPC) drives the frontier of science, finance, engineering, education, chemistry, genomics, modeling, simulation and a host of other compute-intensive industry verticals. The traditional way of running HPC on-premises involves long procurement cycles, high up-front capital investment and, increasingly often, mid-cycle technology refreshes.
The flexibility and heterogeneity of HPC cloud services provide a welcome contrast to the constraints of onpremises HPC. Every HPC configuration is potentially accessible to any given workload in a well-resourced cloud HPC deployment, with vast scalability to spin up as much compute as that workload demands in any given moment.
In 2011, 13% of all sites running HPC leveraged public cloud HPC services. In 2018, the amount rose to 74%. As HPC in the cloud expands into more and more industries and use cases, Amazon Web Services (AWS) is broadening its footprint too. AWS is the primary cloud provider for 58% of the surveyed HPC user organizations running HPC workloads in the cloud, according to Hyperion Research.