[Bucardo-general] Swap replication latency

[Alex Balashov]

On 05/11/2011 10:01 AM, Alex Balashov wrote:
> On 05/11/2011 09:58 AM, David Christensen wrote:
>
>> ISTR you had some user-defined triggers on these tables as well; what
>> types of actions do these triggers perform?
>
> Some business-layer logic leading to minor adjustments of values in
> other tables.
>
> However, this should be irrelevant, if I understood Rosser Schwarz
> correctly here:
>
> https://mail.endcrypt.com/pipermail/bucardo-general/2011-May/000984.html
>
> Because the receiving side is the 'replica', user-space triggers do not
> fire, right?  If they did, that would conflict with row-level deltas
> from the tables those triggers (might, and in my case do) modify.
>
>> Additionally, are thereany rules at play?
>
> None.
>

By the way, I may have underestimated my throughput requirements a bit. 
  I wrote this script to gauge per-second delta changes every 5 seconds, 
between purges:

    http://pastebin.com/BxCtJPWT

And here's what I got:

Delta count: 49003 / 5 second rate: 41.00 deltas/sec
Delta count: 49228 / 5 second rate: 50.60 deltas/sec
Delta count: 49494 / 5 second rate: 54.00 deltas/sec
Delta count: 49710 / 5 second rate: 43.80 deltas/sec
Delta count: 49998 / 5 second rate: 52.00 deltas/sec
Delta count: 50270 / 5 second rate: 55.80 deltas/sec
Delta count: 50485 / 5 second rate: 51.00 deltas/sec
Delta count: 50858 / 5 second rate: 54.80 deltas/sec
Delta count: 51318 / 5 second rate: 95.60 deltas/sec
Delta count: 51786 / 5 second rate: 101.60 deltas/sec
Delta count: 52135 / 5 second rate: 69.40 deltas/sec

-- 
Alex Balashov - Principal
Evariste Systems LLC
260 Peachtree Street NW
Suite 2200
Atlanta, GA 30303
Tel: +1-678-954-0670
Fax: +1-404-961-1892
Web: http://www.evaristesys.com/