I upgraded Pushin’s database from 2 shared CPUs, 2GB RAM to 2 dedicated CPUS, 8GB RAM and holy moly, I did *not* expect it to make such a big difference. HTTP responses sped up 5x (500ms->100ms P95) and queries 2-3x (20-30ms->9.5ms). Running a DB constantly at 80% RAM slows it down a lot it seems.

Managed by Scaleway

Good one! I gotta update mine as well.

I upgraded Pushin’s database from 2 shared CPUs, 2GB RAM to 2 dedicated CPUS, 8GB RAM and holy moly, I did *not* expect it to make such a big difference. HTTP responses sped up 5x (500ms->100ms P95) and queries 2-3x (20-30ms->9.5ms). Running a DB constantly at 80% RAM slows it down a lot it seems.

I've been building something with a few good folk → I promise most of this is JSON, not code

:rest_for_one - our food rest for her

This is how Bella makes sure I finish my work. If I can’t get up I have no choice but to continue working. Very smart supervisor.

Like a painting!

Oh no Mat, you won’t get away this easily 😜

These are the use cases for AI I’m extremely excited and happy about. Somebody had a question but not the means to answer it. So, he used an AI to do the grunt work and found an amazing answer that he would never have found without these insanely powerful tools at his disposal. Absolutely brilliant.

I fixed ~50 security findings reported by Codex Daybreak Blue (their defensive cyber model) in Pushin today. Nothing serious, just a bunch of small DoS vectors. I could have built 3 features in that time, but to be honest: I'm happier this way.

The change is now live! Let me know if your integration breaks and I'll send you a care package.

Thank you :) I'm very happy you like it ❤️

I now have access to Mythos 5.1 RIP our CIs