The task will grow to fill (and exceed) the capacity of the computer available. I worked in a geophysical processing center a while back. We got a CDC 205 vector computer. It was fast. But the FORTRAN compiler stunk. The processors were elated to attack problems too big for the previous computers. They filled it. Then CDC came out with an improved and better optimized FORTRAN. The capacity of the 205 grew. And the size of the problems grew. This continued for many months with everybody happy because the computer appeared to grow and it kept the processors happy because they could take on bigger and bigger problems.
The same thing is happening with PCs and internet. It keeps growing. And with it, the inefficiency of the software. On Sat, May 7, 2011 at 3:06 AM, Dejun Guo <[email protected]> wrote: > With cloud computing (TM hype) more popular, we have to think about > programming efficiency both in size and speed, because it's paid by IO and > time. > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
