News for hackers. Stuff that matters.
Re:Building resilient services at Prime Video with chaos eng... by fxtentacle on Monday August 24, 2020 @03:06PM (#24264207)
If the underlying hardware is a bare metal server, it won't magically turn slow and have a CPU spike. That problem is caused noisy neighbor and kind of exclusive to clouds.

Well, with the 2x example, my app might get from a 1s P99 to a 2s P99 which feels slow, but is still doable. Again, those timeouts are usually introduced by cloud infrastructure. For example, if you use nginx outside of Heroku, it won't have a 30s timeout for file downloads.

Re:Building resilient services at Prime Video with chaos eng... by ses1984 on Monday August 24, 2020 @03:12PM (#24264267)
Your own instances can have an unexpected CPU spike.

Even if you're running on bare metal I find it hard to believe you don't have a layer with short timeouts between your front and backend.

Re:Building resilient services at Prime Video with chaos eng... by fxtentacle on Monday August 24, 2020 @04:16PM (#24265025)
Why would I? I have redundant 1GBit LAN cables between front end, back end, and database servers.
Re:Building resilient services at Prime Video with chaos eng... by ses1984 on Monday August 24, 2020 @06:17PM (#24266165)
Because it's bad ux for your users to see a spinning loading icon forever.
Re:Building resilient services at Prime Video with chaos eng... by fxtentacle on Monday August 24, 2020 @06:36PM (#24266324)
And a timeout error would be better?
Re:Building resilient services at Prime Video with chaos eng... by ses1984 on Monday August 24, 2020 @09:55PM (#24267491)
In my experience, yes, a lot better.