[dnsdist] [EXT] Re: Performance/tuning sanity check

[Remi Gacogne]

On 03/01/2023 17:07, Dan McCombs wrote:
> I spent some time looking through the thread Nicolas shared at 
> https://mailman.powerdns.com/pipermail/dnsdist/2017-April/000281.html 
> <https://mailman.powerdns.com/pipermail/dnsdist/2017-April/000281.html> 
> of his performance tuning experience. Do you know if it's still the case 
> in newer versions of dnsdist to need multiple instances of dnsdist 
> running on a machine this size (32 cores) rather than just additional 
> listeners via addLocal to really push past 400kQPS? I just want to make 
> sure I'm not working off out-of-date information.

A lot has changed with regard to scalability on machines with a great 
number of cores. I remember us doing a lot of work in that area in 
1.3.0, and we also added a few more improvements since then (false 
sharing work-around, more sharding, more efficient locking, etc)

I have personally reached quite more than 400k QPS with a single dnsdist 
process during my tests, but you might be able to get better performance 
with several processes as we are still sharing some state between 
threads. For example Otto has been working on more scalable counters in 
the recursor lately, and this is something that I would like to 
implement in dnsdist. It will likely have to wait until 1.9, so around 
Q3 this year, unless someone beats me to it :)

Please report back with your findings, as it is really helpful to hear 
about different setups than the ones we already do our performance 
testing on.

-- 
Remi Gacogne
PowerDNS.COM BV - https://www.powerdns.com/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.powerdns.com/pipermail/dnsdist/attachments/20230103/90b2bdb5/attachment-0001.sig>