Optimal "set and forget" values?

Is there an optimum "set and forget" value for automatic range identity?
Set it and forget it is best when you know you are never going to add new
subscribers.
When you set it you want to use values which are going to be valid for the
lifetime of your replication solution. So for high throughput solutions, it
could be in the millions; for low throughput solutions it could be in the
10's of thousands.
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
"Earl" <brikshoe@.newsgroups.nospam> wrote in message
news:OyRHOV8aFHA.3280@.TK2MSFTNGP09.phx.gbl...
> Is there an optimum "set and forget" value for automatic range identity?
>
|||I have a merge solution which could have as many as 50 PocketPCs (unlikely)
or as few as 5-6. Right now, I'm using an identity increment of 1 and an
identity range of 1,000,000 for both the publisher and the subscriber. The
typical subscriber would add maybe 200 records (per table) on a day, merge
changes back to the main SQL database, then delete the subscription at the
client. So I suppose that this really becomes an unlimited number of
subscribers, since a new subscription is sent down every day? I'm trying to
forecast the future here and it seems pretty murky to me. Thanks for the
advice.
"Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
news:Oi9ZZn8aFHA.1148@.tk2msftngp13.phx.gbl...
> Set it and forget it is best when you know you are never going to add new
> subscribers.
> When you set it you want to use values which are going to be valid for the
> lifetime of your replication solution. So for high throughput solutions,
> it
> could be in the millions; for low throughput solutions it could be in the
> 10's of thousands.
> --
> 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
> "Earl" <brikshoe@.newsgroups.nospam> wrote in message
> news:OyRHOV8aFHA.3280@.TK2MSFTNGP09.phx.gbl...
>
|||you should be good for 2,147,483,648/1,000,000 -1 (to account for the
publisher) or roughly 2,146 subscribers. If all goes well and if you expect
less than 1 million rows to be added on your publisher and each subscriber
for the lifetime of your replication solution.
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
"Earl" <brikshoe@.newsgroups.nospam> wrote in message
news:edYmcA%23aFHA.2288@.TK2MSFTNGP14.phx.gbl...
> I have a merge solution which could have as many as 50 PocketPCs
(unlikely)
> or as few as 5-6. Right now, I'm using an identity increment of 1 and an
> identity range of 1,000,000 for both the publisher and the subscriber. The
> typical subscriber would add maybe 200 records (per table) on a day, merge
> changes back to the main SQL database, then delete the subscription at the
> client. So I suppose that this really becomes an unlimited number of
> subscribers, since a new subscription is sent down every day? I'm trying
to[vbcol=seagreen]
> forecast the future here and it seems pretty murky to me. Thanks for the
> advice.
>
> "Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
> news:Oi9ZZn8aFHA.1148@.tk2msftngp13.phx.gbl...
new[vbcol=seagreen]
the[vbcol=seagreen]
the[vbcol=seagreen]
identity?
>
|||Thanks Hilary. My expectation is for less than 1 million rows on both
publisher and subscribers. One thing that still concerns me however -- when
I delete on-board databases (subscribers) and create anew, doesn't this
indeed create a new subscriber and eat up one of my "2,146 subscribers"?
"Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
news:%23enx9fBbFHA.3840@.tk2msftngp13.phx.gbl...
> you should be good for 2,147,483,648/1,000,000 -1 (to account for the
> publisher) or roughly 2,146 subscribers. If all goes well and if you
> expect
> less than 1 million rows to be added on your publisher and each subscriber
> for the lifetime of your replication solution.
> --
> 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
> "Earl" <brikshoe@.newsgroups.nospam> wrote in message
> news:edYmcA%23aFHA.2288@.TK2MSFTNGP14.phx.gbl...
> (unlikely)
> to
> new
> the
> the
> identity?
>
|||yes, when you create a new subscriber, whether the pda had an old one before
or not, the max identity value on the publisher will be updated. This max
value is consulted when incrementing an existing range or assigning a new
one.
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
"Earl" <brikshoe@.newsgroups.nospam> wrote in message
news:%23hY8tbHbFHA.3848@.TK2MSFTNGP10.phx.gbl...
> Thanks Hilary. My expectation is for less than 1 million rows on both
> publisher and subscribers. One thing that still concerns me however --
when[vbcol=seagreen]
> I delete on-board databases (subscribers) and create anew, doesn't this
> indeed create a new subscriber and eat up one of my "2,146 subscribers"?
> "Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
> news:%23enx9fBbFHA.3840@.tk2msftngp13.phx.gbl...
subscriber[vbcol=seagreen]
an[vbcol=seagreen]
trying[vbcol=seagreen]
the[vbcol=seagreen]
for
>
|||Very helpful Hilary. Now I just need to make a minor change to my app so as
to ensure I do not have any long-term issues with the number of subscribers.
I've estimated that -- in most scenarios -- I would've run out of identity
ranges in about a year otherwise.
"Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
news:%23YPrULJbFHA.2380@.tk2msftngp13.phx.gbl...
> yes, when you create a new subscriber, whether the pda had an old one
> before
> or not, the max identity value on the publisher will be updated. This max
> value is consulted when incrementing an existing range or assigning a new
> one.
> --
> 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
> "Earl" <brikshoe@.newsgroups.nospam> wrote in message
> news:%23hY8tbHbFHA.3848@.TK2MSFTNGP10.phx.gbl...
> when
> subscriber
> an
> trying
> the
> for
>