Keep in mind the AMD instances are using chips AMD put out almost 3 years ago. They still aren't using the new generation of Rome chips that came out a year ago, which almost doubled performance.
We’re excited to start evaluating AWS’ ARM offerings, but until now, they would’ve been more expensive for our workload. We’ll have to see if that’s changed with Graviton2, but I suspect AMD is the sweet spot for our application; it just doesn’t seem to like ARM as much.