Maintained by: NLnet Labs

[Unbound-users] Unbound multithread performance: an investigation into scaling of cache response qps

W.C.A. Wijngaards
Wed Mar 24 11:29:08 CET 2010


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Aaron,

Another check with do-tcp: no, so that the evport system calls do not
have to lock that file descriptor to inspect if events happened on the
TCP fd.

On 03/24/2010 11:08 AM, W.C.A. Wijngaards wrote:
> A control check, with four perfs running towards unbound.
> evport, forked, 4senders:  9619  15860  19010  21979
> evport, forked, 2senders:  9700  17300  19600  22300
> 
> evport, forked, special:   10000  18783  23461  25797

evport, forked
special, no-tcp :            10200  18552  20226  27161
again                                      23646  27149
and again                                  19350  27792

The 3-core output is very bouncy. The four-core throughput is more stable.

So, that removes another lock, and speeds up a bit more :-)

Best regards,
   Wouter
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/

iEYEARECAAYFAkup6XQACgkQkDLqNwOhpPgzWwCeOLuVVAOVd7JVDVjdwP7DiYZA
c2EAoIUPb/T2j7eAb5P7dQhHpH6T/wWn
=peg8
-----END PGP SIGNATURE-----