Re: Article about supposed "murky" future for Oracle

"Serge Rielau" <srielau_at_ca.eye-be-em.com> wrote

> Wouldn't you have some sort of a cut off time? I mean you would look at
> all transactions that happened for a given time-period using a timestamp.
> If the DBMS kept running sums in the operational system then it buys
> itself a "writers blocking writers" problem in the normal operations and
> it wouldn't be normalized either.
> As a result the data that is being looked at is stable. RR/Serializable
> (if chosen) ensures no-one is trying to undo a previously committed
> transaction but it does not harm incoming new transactions.
> Maybe I'm naive....

the other way to put this is that application has to be designed accordingly. In a properly designed system, such exaggerated scenarios of read-blocking-writes rarely happens or can be completely eliminated.

This is what I call as marketing hype. Present a scenario about competing products as if it is insurmountable.
Of course Oracle is not the only one to do it. IBM seems to be learning fast, as evident in this

http://www-3.ibm.com/software/data/pubs/papers/readconsistency/readconsistency.pdf Received on Wed Mar 31 2004 - 16:11:56 CST