Re: openldap.git branch mdb.master updated. 0ab841598ffb490f4246f892248f0b409e411cc1

[masarati@aero.polimi.it]

> Howard Chu wrote:
>>> - Log -----------------------------------------------------------------
>>> commit 0ab841598ffb490f4246f892248f0b409e411cc1
>>> Author: Howard Chu<hyc@symas.com>
>>> Date:   Sun Sep 18 16:39:18 2011 -0700
>>>
>>>       Fix 09006ccec7928c9cf53bca6abe741e8d4d466c98
>>>
>>>       Check for stale DBs was in the wrong place.
>>
>> Since back-mdb does no caching, it's OK to run slapadd while slapd is
>> running;
>> slapd will see the new data immediately. In fact you can run multiple
>> slapds
>> on the same database and they will stay perfectly in sync. I'm not sure
>> that
>> that's actually a useful thing to do, but you can do it if you want...
>
> To answer my own question - apparently it's useful to get around
> bottlenecks
> in the slapd connection manager. Over 123,000 searches/second using two
> slapds
> on the same mdb database. At this point core#0 was at 99.5% busy in soft
> interrupts. The slapds were only using (combined) 1250% CPU.
>
> http://highlandsun.com/hyc/slamd-mdb/jobs/job_20110918191657-463348391.html

... which goes back to LDAP's original purpose: lots of reads, occasional
writes.  Sounds like back-mdb is LDAP's holy graal :)

p.