Issue link: https://read.uberflip.com/i/1187300
Amazon Web Services – HPC Lens AWS Well-Architected Framework Page 32 • Selection • Review • Monitoring • Tradeoffs The review, monitoring, and tradeoffs categories will not be described in this paper because the best practices in the AWS Well-Architected Framework whitepaper do not require modification for HPC workloads. Refer to the AWS Well-Architected Framework paper for an understanding of best practices. Best Practices Selection The optimal solution for a particular system will vary based on the kind of workload you have. Often there are multiple approaches possible. Well- architected systems use multiple solutions and enable different features to improve performance. An HPC architecture can rely on one or more different architectural elements, for example, queued, batch, cluster compute, containers, serverless, and event-driven. Compute The optimal compute solution for a particular HPC architecture depends on the workload deployment method, degree of automation, usage patterns, and configuration. Different compute solutions may be chosen for each step of a process. Selecting the wrong compute solutions for an architecture can lead to lower performance efficiency. Instances are virtualized servers and come in different families and sizes to offer a wide variety of capabilities. Some instance families support specific workloads, for example, compute-, memory-, or GPU-intensive workloads, while others are general purpose. Both the targeted workload and general- purpose instance families are useful for HPC applications. Instances of particular interest to HPC include the compute-optimized family and accelerated instance types such as GPUs and FPGAs. However, compute- intensive workloads will see significant performance degradation on the T- series instance family, even with Unlimited mode enabled, since its compute resources are designed for applications with moderate CPU usage that experience temporary spikes in usage.