named taking up 100resource on only 1 CPU on a multi CPU sy stem

Simon Waters Simon at wretched.demon.co.uk
Sat Jan 11 21:47:58 UTC 2003


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

Kuhtz, Christian wrote:
>
> While I agree that BIND9 isn't as quick as BIND8, the solution
for running
> BIND8 on multiproc is IMHO ugly enough operationally for some
of us to cause
> us to want to bite the bullet of BIND9 now, and hope for
performance
> improvements as we move forward.

Rick's own figures show a case where roughly 2 CPU's with BIND 9
match one CPU's with BIND 8, so a box with four CPU's should
easily out perform a single BIND 8, more a question of does the
OP want to commit the extra resource to DNS in order to run BIND 9.

My guess also is Rick's testing is slightly idealised, I'm
guessing even raw HP internal DNS is a lot cleaner than the
Internet DNS (either that or the DNS luminaries to escape from
HP must have left in disgust <hehe>).

BIND 9 is less keen to restart queries in pathological
situations where zones having incorrect name server information,
and so I'd be curious how the numbers stack up is a real world
test, but appreciate the difficulty involved in performing
these. My guess is BIND 9 will close the gap slightly, but
perhaps something other than the delegation handling dominates,
or it may be too small to be noticable.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+IJELGFXfHI9FVgYRAvDxAJ9nBAnFBRiRT9oBPifpyattV5sCPQCgjxZh
MW1EUOm0XM/6StV3KyBtpnY=
=JN45
-----END PGP SIGNATURE-----



More information about the bind-users mailing list