Merge Replication Speed

Hello,
When I first started experimenting/testing Merge Replication about a year
ago, I found that the default Merge Replication speed seemed to be a full
minute. After doing some research, I tried creating a profile and changing
the -PollingInterval parameter from 60 to 1. This got changes moving across
about once per second (as expected.) Now, when I'm running the same tests,
I'm finding that the maximum speed seems to be once every five seconds, even
though I have again changed the -PollingInterval to 1 second.
Has something changed in a service pack in SQL where replication will go no
faster than once every five seconds? Any information would be greatly
appreciated!
Thanks, and thanks for all the help others have given me through this group
to date! It has not been insignificant!
Marshall
Merge replication's latency on a high throughput system is typical between 1
to 2 minutes. It is possible that there is more metadata in the merge
replication tables which is causing the latency to be larger than the 1 s
that it was before.
AFAIK there has been no significant change to the merge replication process
which would account for this increase in latency.
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Marshall Cummings" <mbcdev@.yahoo.com> wrote in message
news:1118427207.284445@.nnrp1.phx1.gblx.net...
> Hello,
> When I first started experimenting/testing Merge Replication about a year
> ago, I found that the default Merge Replication speed seemed to be a full
> minute. After doing some research, I tried creating a profile and
> changing
> the -PollingInterval parameter from 60 to 1. This got changes moving
> across
> about once per second (as expected.) Now, when I'm running the same
> tests,
> I'm finding that the maximum speed seems to be once every five seconds,
> even
> though I have again changed the -PollingInterval to 1 second.
> Has something changed in a service pack in SQL where replication will go
> no
> faster than once every five seconds? Any information would be greatly
> appreciated!
> Thanks, and thanks for all the help others have given me through this
> group
> to date! It has not been insignificant!
>
> Marshall
>