|
I’m seeing similar issues: create a
CFC (or several) that work together and all is well – but use them
several times and performance tanks. In my case I’ve got a “Festival”
CFC that contains many “Venues” (about 60), “Events” (about
100), “Artists” (about 100) and “Schedules” (which link
events to Venues at particular times).
When this is empty everything runs fine. Add in all the information and
performance get really loopy.
Sometimes fully populating the festival (which is then cached in the
Application scope) takes as little as 8 seconds – other times as long as
2 minutes. This is on a completely dedicated
development system with no (observable) other load (at least nothing that’s
appearing in PerfMon) and a single user. In the debug output everything will fly then
I’ll see one otherwise simple method call (something like a Venue.new())
taking 70 seconds or a series of simple calls (something like venue.getparent() which runs once for each venue and
normally averaging 0ms taking 1200ms each). I admit this may be an issue with servers…
but I see the exact same behavior on my hosted production server and on another
box I set up to test. That being said even when I’m not
seeing inexplicable performance dips CFCs definitely don’t seem to scale
linearly – or, to be fair, at least MY CFCs don’t. ;^) In general I guess all I’m saying is
make sure to test them both under load and with as many instances as you’ll
be using in production. And it anybody has any suggestions for me
to tame my beast I’m all ears! -----Original Message----- I have a CFC that processes sales orders. It has a
Customer.cfc and Product.cfc to validate each of these properties. On my dev
machine, it took 591ms to create and save 1 order with 5 items. Should I
be satisfied with this performance or should I look at ways to improve
this? The majority of this time was in the 5 calls to the
AddItem method which has a product CFC verify item information, with the
first call taking 80ms, then 50ms, 50ms, 30ms and 30ms. I am using an Access DB and am in the process of moving to
MS SQL 2000. Should this improve performance? Andy |
- RE: [CFCDev] CFC Performance question Andy Ousterhout
- RE: [CFCDev] CFC Performance question Joe Eugene
- RE: [CFCDev] CFC Performance question Andy Ousterhout
- Re: [CFCDev] CFC Performance question Matt Liotta
- Re: [CFCDev] CFC Performance questio... John Farrar
- Re: [CFCDev] CFC Performance que... Matt Liotta
- RE: [CFCDev] CFC Performance questio... Andy Ousterhout
- Re: [CFCDev] CFC Performance que... Matt Liotta
- RE: [CFCDev] CFC Performance que... Nat Papovich
- RE: [CFCDev] CFC Performance que... Robin Hilliard
- RE: [CFCDev] CFC Performance question Jim Davis
- RE: [CFCDev] CFC Performance question Joe Eugene
- RE: [CFCDev] CFC Performance question Jim Davis
- RE: [CFCDev] CFC Performance question Nando
- RE: [CFCDev] CFC Performance question Jim Davis
- RE: [CFCDev] CFC Performance questio... Nando
- RE: [CFCDev] CFC Performance que... Tim Blair
- RE: [CFCDev] CFC Performance que... Jim Davis
- RE: [CFCDev] CFC Performance que... Tim Blair
- RE: [CFCDev] CFC Performance que... Jim Davis
- RE: [CFCDev] CFC Performance que... Tim Blair
