Can you tell me which is more efficient in easytrieve.

prash_infy · New User Joined: 31 Aug 2006 Posts: 20

Want to know which is more efficient in performance and why.

dick scherrer · Posted: Mon Jun 14, 2010 9:12 pm

Hello,

One way to be more efficient is to make sure that the condition most likely to be true is specified first. . .

You could also run some timing tests but you would need to have a few million records . . .

prash_infy · New User Joined: 31 Aug 2006 Posts: 20

Thanks d.sch

Of the two codes mentioned above, which one do you think is more efficient?

dick scherrer · Posted: Mon Jun 14, 2010 9:24 pm

Hello,

As i mentioned, the one that looks at the value most likely to be true first will be more efficient.

Either format will probably run about the same, so you want to eliminate as many "false" tests as possible to reduce the amount of internal processing.

The only way you can see this is to run some very large tests. A small test will show very little or no difference.

dbzTHEdinosauer · Posted: Mon Jun 14, 2010 10:00 pm

prash_infy,

what Dick is trying to get you to do is
consider the impact of
IF WS-VAR2 > 0 or WS-VAR1 = 'A'
versus
IF WS-VAR1 = 'A' or WS-VAR2 > 0

are there more WS-VAR2 situations
than WS-VAR1?

the first condition, if true, will cause immediate transfer
otherwise
the second condition is checked.

So if you need to determine if there is statistically
(based on analysis of input data)
a higher chance of one over the other occuring;
the statistically higher should be coded first.
This is also referred to as fast track.

prash_infy · New User Joined: 31 Aug 2006 Posts: 20

Thanks Mr Brenholtz!!!!