Re: Synchronized Applications

From: Mike Monagle (mikemonagle_at_nospam.comcast.net)
Date: 07/28/04 

Date: Wed, 28 Jul 2004 09:40:07 -0600

I believe SQL Server calls the first method you described "log shipping" (in
combination with a transaction queue). This definitely is an option. I
read somewhere that the database is inoperative while it is importing logs.
I don't know if this will fly. Hopefully we'll have time to build some
prototypes before making our final selection! Management wants every server
to be 100% in synch. at all times. Don't know that this is achievable with
any method. We'll see...

I'll look into Oracle self-synchronizing databases. We have tried SQL
Server replication and found that it takes hours for the database to get in
synch (while generating a lot of network traffic).

Thanks! Mike

"Roedy Green" <look-on@mindprod.com.invalid> wrote in message
news:9pidg0dn263sr1494ef68nqdfbo8igj4bp@4ax.com...
> On Tue, 27 Jul 2004 09:56:56 -0600, "Mike Monagle"
> <mikemonagle@nospam.comcast.net> wrote or quoted :
>
> >Use of message queues, replicated databases, other third party solutions,
> >and home-grown solutions are all options.
>
> Probably the simplest way to handle it is to make mods to the local
> database via a transaction queue. You can think of it is a list of
> records, or list of object that describe changes to make to the
> database. Only things that potentially change the database are
> included.
>
> Then you log these for backup. If you have trouble, you can replay the
> transactions against an old snapshot of the database. It takes a
> while to recover, but the overhead of logging is minimal.
>
> You can send the log to the other two stations, who can replay it
> ongoingly to keep a copy hot backup copy of your database going.
>
> This assumes it that precise sync is not neeed. One database can be
> several seconds behind.
>
> Oracle has self synching databases. They don't require you to think
> totally in terms of transaction, but they don't aim for up to the
> second sync either.
>
> You might need to timestamp each transaction, broadcast them, and only
> process the transaction once you have notification from each station
> of the transactions for that second, or that they have none. I am
> speaking roughly.
>
>
>
> --
> Canadian Mind Products, Roedy Green.
> Coaching, problem solving, economical contract programming.
> See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.

Relevant Pages

  • Re: transactional replication problem
    ... All transaction from killed process were rolled back... ... Did they get deleted from distribution database ... >Looking for a SQL Server replication book? ...
    (microsoft.public.sqlserver.replication)
  • Re: Help - Timing Logic
    ... server application, both of which ran on the same box. ... the client applications and 'lodging' them in the database. ... Another part of the server application was dedicated to retrieving messages ... commit transaction ...
    (microsoft.public.dotnet.languages.vb)
  • Re: Help - Timing Logic
    ... put a trigger on the database table. ... Transaction and Database Locking - look at isolation levels / settings ... Maybe implement a message broker ... ...
    (microsoft.public.dotnet.languages.vb)
  • Re: Help - Timing Logic
    ... build a trigger on the database .. ... Transaction and Database Locking - look at isolation levels / settings ... If you need to implement a locking mechanism / or / logging mechanism / or / a checking mechanism to avoid duplicate messages caused by multi-threading ... ... message broker gets all the necessary messages or message id's from the database to be sent ... ...
    (microsoft.public.dotnet.languages.vb)
  • Re: Help - Timing Logic
    ... put a trigger on the database table. ... Transaction and Database Locking - look at isolation levels / settings ... Maybe implement a message broker ... ...
    (microsoft.public.dotnet.languages.vb)