Mike Austin discusses balancing performance and cost in development environments. He focuses on storage management—questioning the need for expensive SSDs in dev when spinning disks may suffice—and on committing to reserved instances (short vs long term commitments) after understanding application needs. The segment emphasizes data-driven decisions about environment configuration to reduce costs without harming production performance.
Chapters
How do we balance performance and cost? Right. You know, I can give you an example of something maybe not quite as effective here, but probably effective. Storage management more so. But so often we'll see dev environments that have SSD drives which are very expensive. So why, why do you have SSD drives in the dev environment? And a lot of times the engineering architectural teams will say, well, because we mirror production into dev and it's like, okay, cool, but do you really need to have SSD in the dev environment? Like if you just spinning, spinning disk and its performance is slower in the dev environment? As long as it performs, you know it's going to perform when you go to ssd. Right now, I'm probably oversimplifying, but you kind of get the picture, right? Those are the sorts of things you want to be looking for. You definitely want to be looking at your reserved instances. You need to have the data to figure those out, right? So do we make the long term commitments? Because three years are, you know, more cost effective than one year. But I have to understand the environment, I have to understand the applications, I have to understand what's going on in order to make those Sam.