Re: TPC-A and -B Benchmarks

Jonathan Leffler writes: |> The issue rapidly reduces to: what does who mean by repeatable read. |> Oracle supplies one definition; Informix supplies another; I can't |> usefully comment on 3rd party products. The two definitions differ, |> which is very unfortunate. Jonathan, I hate to diagree with you so vehemently in so public a forum, but I fear you have forced my hand. What is meant by "repeatable read", the definition used, is not at the whim of any db vendor. The theory and concepts of isolation levels have been "defined" by the likes of Date (actually, Date credits J.N.Gray, R.A.Lorie, G.R. Putzolu, and I.L.Traiger for first describing the concept of isolation level in their article "Granularity of Locks and Degrees of Consistency in a Shared Data Base" {Database Systems, Vol 1, 4th Ed - C.J.Date, pg. 433}), and the SQL Standard. What differes among the db vendors is the method they have chosen for implementing these features. In this respect, it is quite easy to determine if a given vendor has implemented the feature as it is defined. There is no differing of definitions. Now I suppose you could argue against having to subscribe to those definitions, but given the general adherence to the SQL standard by so many db vendors, such an argument would be little more than an intellectual exercise. Dave Disclaimer: These opinions are not those of Informix Software, Inc. ************************************************************************** "I look back with some satisfaction on what an idiot I was when I was 25, but when I do that, I'm assuming I'm no longer an idiot." - Andy Rooney