Re:Large-scale, semi-automated Go GC tuning
by yxhuvud on Wednesday January 12, 2022 @06:28AM
(#29904953)
Makes me wonder what on earth all those cpus are doing. Many people take trips by taxi, but not THAT many.
[ Parent
| on HN ]
Re:Large-scale, semi-automated Go GC tuning
by donatj on Wednesday January 12, 2022 @06:59AM
(#29905179)
Right? Like our product isn’t Uber scale, but our Go microservices each get about 10,000 active users per core before the autoscaler adds a new instance. I’m curious their numbers.
Re:Large-scale, semi-automated Go GC tuning
by zimpenfish on Wednesday January 12, 2022 @07:42AM
(#29905555)
"Uber drivers completed 4.98 billion trips in 2020, a 27 percent decrease from the 6.9 billion trips in 2019"
Now, granted, 2021 is probably down on that further but it's still a lot.
Re:Large-scale, semi-automated Go GC tuning
by speedgoose on Wednesday January 12, 2022 @07:57AM
(#29905729)
7 billion per year is 222 trips per second on average.
Re:Large-scale, semi-automated Go GC tuning
by christophilus on Wednesday January 12, 2022 @08:43AM
(#29906167)
With each trip taking an average of 15 minutes, that’s about 200k concurrent, live sessions. Each session has at least two participants, so 400k live users any given second. I imagine their spikes are significantly higher than their slow periods, too.
Re:Large-scale, semi-automated Go GC tuning
by yxhuvud on Wednesday January 12, 2022 @01:53PM
(#29911238)
Well sure. And I've worked on software run by ISPs that have had more than 4M concurrent active sessions. Handled by something like 20 machines. Yes, they probably have way more overhead per session, but it is still a really high amount of servers.
Re:Large-scale, semi-automated Go GC tuning
by speedgoose on Wednesday January 12, 2022 @10:12AM
(#29907408)
So a bit less than 6 live connections per CPU saved with the garbage collector configuration.