• Jason A. Long (5/21/2015)


    Ed Wagner (5/21/2015)


    Jason A. Long (5/21/2015)


    MikeTomkies (5/21/2015)


    I have implemented Jason solution and understood it after reading the article thank you Jason. I see luis has improved it which i will give it a try tomorrow and digest.

    Thank you guys for your help you have really taught me loads.

    For the record Jasons Solution took 15485 stock records split them out into 38626 Rows in under a second. Something i did not think possible until today.

    Luis' solution should be even faster. If the estimated execution plan is an indicator (not always the case), his should be much faster.

    In either case, the "magic sauce" they both have in common is DelimitedSplit8K function...

    I find that function has the tendency to change the way people look at data and their expectations of performance after using it a few times. 😉

    I think that's the polite way of saying, "It completely destroys other SQL based splitters"...

    Yes it does. However, it does more than that. The techniques the article teaches and the approach to testing can be used to make many other things more efficient. It's also fun. 😉