search

Netflix / Hystrix

Hystrix is a latency and fault tolerance library designed to isolate points of access to remote systems, services and 3rd party libraries, stop cascading failure and enable resilience in complex distributed systems where failure is inevitable.
24.08k stars 4.7k forks source link

Hystrix Thread Pool #1991

Open karunagoyal opened 4 years ago

karunagoyal commented 4 years ago

Hello,

I am working on a micro service in which I am using Hystrix command with THread isolation strategy to make network call. My network latency is very high with TP99 as high as 3 second or more. My Request Per second (tps) can go up to 60 or more. Based on recommended formula "requests per second at peak when healthy × 99th percentile latency in seconds + some breathing room" my core pool size come very high number. Can you please suggest the configuration fir this use case? Or if there is a better option.

karunagoyal commented 4 years ago

Can someone please share their experience?

GiteshKhannaOYO commented 4 years ago

Any issues that you are facing with such a high thread pool?

Also, this formula actually seems to be coming taking into account the worst-case scenario which is that every request is being served at P99. I think you can get away with something smaller than that if the requests being served at P99 are actually very minimal. Just keep the execution timeout high.