El sáb., 20 abr. 2019 a las 0:05, Peter Booth via nginx
(<nginx@nginx.org>) escribió:
>
> Where is your upstream? Where is your pho executing? Do you have a CDN?

Hi Peter , both my upstream and php application are in the same
datacenter ,different virtual machines with kvm, 1gb network , I do
not have CDN

>
> There’s three parts to this:
> 1 fix the bad OS defaults:
> If you are using RHEL 6 this would mean:
> Enabling tuned
> Disabling THP
> Increasing vm.min_free_kbytes
> Reducing swappiness

Ok , here I understand, the difference is that I'm using Centos 7

> 2 generic web server specific configuration

I do not understand, you need the specification of each server? ,
each server runs on a virtual machine with kvm, 4 cores xeon 2.1Ghz
with 4gb ram, disk 15k

> Increasing # ephemeral ports

I do not understand, could you please detail more.

> Adjusting ulimit

ok

> 3 tuning specific to your workload.
> UDP and TCP buffet size - should match BDP

please specify a little more.

> NIC tuning - irq coalescing, 10G specific tuning, see CDN, melanox,redhat, hp suggestions for lie latency tuning.
>
> That’s a start.

Ok .

--
rickygm

http://gnuforever.homelinux.com
_______________________________________________
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx