On Mon, Aug 27, 2007 at 07:07:58PM -0600, Alvaro Mantilla Gimenez wrote:
> Jacob Yocom-Piatt wrote:
>> tried to take a bit of a side adventure and get HP-UX going on a PA-RISC 
>> machine and it's no walk in the park. for cost, support, compatibility and 
>> simplicity reasons i've abandoned the project and decided to use other 
>> OSes instead.
>
> How was your adventure?? Can you be more specific?? I know the cost 
> part...obviously it is more cheaper run OpenBSD that HP-UX. But i need 
> more...something really heavy like "I tried to install an OpenLDAP with 
> HP-UX and the system load with 2000 users rise to the sky...but the same  
> number of users with OpenBSD had an incredible performance and never pass 
> from 10% of load"....or whatever...

I wrote a huge mail, but essentially: `what Jacob said'.

I think that if you manage to convince the right people that a network
of smaller nodes has advantages (higher availability, better
scalability, lower `TCO' - whatever), OpenBSD - with low cost, good and
cheap support, and very competitive performance - becomes a very
attractive option.

On the other hand, if you really need a 16-core machine with 32 GB of
memory, installing OpenBSD gets you cool dmesg pr0n but not a really
useful configuration - OpenBSD doesn't perform too well on such beasts.
(Although I see Jacob's point in that you can still use OSS even if it's
not OpenBSD.)

I'd focus my research on this issue.

In the end, though, the goal should not be to use what you like best;
the goal should be to pick what is best for the business. (And OpenBSD
can be a very good fit, and picking something that is not too offensive
to the person maintaining it *is* good for the business; but still.)

                Joachim

P.S. One more issue: you *do* realize that getting OpenBSD to
authenticate against LDAP is not entirely trivial, right? This might be
a serious problem if the LDAP system is to handle network-wide logins...

-- 
TFMotD: dbmmanage (1) - create and update user authentication files in
DBM format

Reply via email to