After experiencing the annoyance of extensive benchmarking while looking for optimal placement for production workloads in a cost-constraint situation, we put t

After experiencing the annoyance of extensive benchmarking while looking for optimal placement for production workloads in a cost-constraint situation, we put together tools to save us time and effort. We kinda feel like this might be useful for others, but a bit confused about potential use cases. So we put together a simple app for others to play with (first use case being benchmarking for AWS SageMaker). For this to work there is a couple of lines of code to bake telegraf-based monitoring into container, script to set-up an IAM role (so app can get this launched under your account safely), then you just pick multiple instances to run the container and get a summary of launches with $, t, links to Grafana with monitoring data and python logs for each launch. It’s free - we plan to make it an open source tool if it proves to be useful, but the web app was something we could put together the fastest. We’d love to hear your feedback - if that’s something that might be of any use?