Grid computing on AWS

A key driver for the migration of HPC workloads from on-premises environments to the cloud is flexibility. AWS offers HPC teams the opportunity to build reliable and cost-efficient solutions for their customers, while retaining the ability to experiment and innovate as new solutions and approaches become available.

HPC teams that want to migrate an existing HPC solution to the cloud, or to build a new solution, should review the AWS Well-Architected Framework which also includes a specific Financial Services Industry Lens with a focus on how to design, deploy, and architect financial services industry (FSI) workloads that promote resiliency, security, and operational performance in line with risk and control objectives. This framework applies to any cloud deployment and seeks to ensure that systems are architected according to best practices. Additionally, the HPC-specific lens document also identifies key elements to help ensure the successful deployment and operation of HPC systems in the cloud.

The following sections include information about AWS services that are most relevant to HPC systems, particularly those that support financial services customers.

A typical HPC architecture with the key components, including the risk management system (RMS), grid controller, grid brokers, and two compute instance pools