mludvig
commented
2 years ago how do I monitor the stack(s) to know that I'm utilizing the most of the available hardware per region?
You can open the EC2 console and in the Instances tab see how many you're running. Since you've got only 8 vCPU service limit you will only be able to run max 2 instances, all the smallest G-class instances are 4 vCPU minimum.
If you've set a high enough required hash rate in the CloudFormation stack parameters (e.g. 1000 MH/s) and you get higher limits approved later on the AutoScaling group will automatically adjust and add more capacity. You don't have to redeploy.
Hi Michael, thanks for putting this together: the templates and the write-up on medium. I'm running your app to try ethmining before it, like btc, becomes out of reach for cloud computing.
It appears that as of July 2022 the lowest spot prices for g4ad are in the London region.
I've made the mistake of performing the first run without increasing the quotas, then asked for large numbers in the one region I was looking at. I got approved for 4, now for 8. I also did not realize the requests have to be made per region, so I just mass-filled out those.
My question is, how do I monitor the stack(s) to know that I'm utilizing the most of the available hardware per region? Currently I don't have much visibility into whether all of the permitted spot instances are being utilized. I guess the dashboard shows workers per region grouped by region codes and the spike in the MH can be attributed to increased capacity, but it's largely guesswork.
Also, I'm assuming that I have to restart the process to pick up additional instances after an increase in quota?
Cheers