Was just looking through older threads (this one's not so old), and found this:
Hm, if I'm contracted out to make a new application, why would I write it with any known future performance problems? The best way to fail to get a followup contract would be for a brand new application I wrote to start choking as more real users are allowed to touch it. I know there's such a thing as premature optimization, but I also know that there's a reason that DBAs get $120K to start... in Kansas.
"Halley, I know we said we expected Frobnitz.com to get a thousand hits a week, but Paris Hilton just mentioned how she loves her new Frobnitz in the middle of her latest inadvertent sex tape. The machine keeps falling over, and we're only getting a thousand hits an hour. Yet our popular Squonk.com site on an identical machine in the same rack is handling five thousand concurrent users easily. What the frell?"
--
[ e d @ h a l l e y . c c ]
In reply to Re^4: Strange behavior: Class::DBI with CGI::Application
by halley
in thread Strange behavior: Class::DBI with CGI::Application
by johnnywang
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |