Re: Phantom Inserts

In article <tIxKl0iA$zTC065yn@login.dknet.dk> zeppelin@login.dknet.dk (Thomas B. Pedersen) writes: >In article <pwongD4C773.F15@netcom.com>, Philip S. Wong wrote: > >> I was wondering how different databases/vendors implement protection >> against phantom inserts. In particular, how can they achieve phantom >> protection while maintaining a good degree of concurrency? > >The ORACLE7 Concepts Manual (10-6) mentions something called a SCN >(system change number) which is used to ensure that transaction A >never sees data committed by transaction B after transaction A started. >ORACLE calls this "statement-level read consistency". > >When providing read-consistency, the manual says, "blocks with changed >data (more recent SCNs) are reconstructured using data in the rollback >segments...". I don't doubt that it is necessary to provide some kind >of read-consistency, but doesn't this approach cause an enormous >performance degradation? Especially for long-running transactions the >same data could have been changed a number of times. > Yes, but if you factor in the complicated buffering algorithms, such as the possibilty that continuing to find the record in memory avoids disk hits, the performance _can_ be much better than one would expect. On the other hand, it can be unexpectedly worse. It is a non-linear function that can only be evaluated empirically. And extremely sensitive to design and tuning. > >zeppelin > > >-- > ____________ ___ ____ _ _ _ _ > |_ | _/| _ | _ | _/| | | | \\| | zeppelin@login.dknet.dk > / /| _>| _| _| _>| |_| | | > /___|___\\|_| |_| |___\\|___|_|_|\\_| -- Joel Garry joelga@amber.rossinc.com Compuserve 70661,1534 These are my opinions, not necessarily those of Ross Systems, Inc. %DCL-W-SOFTONEDGEDONTPUSH, Software On Edge - Don't Push. panic: ifree: freeing free inodes...