Thank you both for all your help. This board would be
sorely lacking without the two of you.
Hilary: Heres the deal. Your custom sync object was
working, mine wasn't.
@.sync_method = N'character'
That was the whole problem. I know you asked me to try
this a while back and I did, but I have no idea what else
I was doing back then that would have affected this
outcome. You're example on the 08/11 didnt specify (not
blaming you by any means) and by then I had forgot you're
mentioning it.
Paul: from you're reply @. 1:59 am. Do you ever sleep?
>
I personally would prefer to use indexed views.
I suppose simplistically the downside is that there is
increased overhead on
the publisher to maintain the materialized data.
I see the huge benifit being not needing to write the
ins,upd, and del procs with the Indexed View option. What
increased overhead are you refering too? Does the whole
Arithabort thing ever become a real pain?
Thanks again both of you.
ChrisR
ChrisR,
for an indexed view you've got 2 types of overhead:
a.. The disk space taken up by the view
a.. The cost of maintaining the view on disk as the base tables are
modified - this is conceptually like a set of invisible triggers.
BTW I'm not really such an SQL insomniac :-) I guess the post times are
Western US timezone times: UK - 7 hours.
Regards,
Paul Ibison
|||Ahhh. I forgot you were in the UK. Thanks.
>--Original Message--
>ChrisR,
>for an indexed view you've got 2 types of overhead:
>a.. The disk space taken up by the view
>a.. The cost of maintaining the view on disk as the base
tables are
>modified - this is conceptually like a set of invisible
triggers.
>BTW I'm not really such an SQL insomniac :-) I guess the
post times are
>Western US timezone times: UK - 7 hours.
>Regards,
>Paul Ibison
>
>.
>
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment