I get this:

"Our Programs must eat less CPU cycles. We write web based software
which runs on servers and servers have to handle lots of processes. We
can put more RAM into the server but CPU cycles are a problem. So our
programs must be efficient in terms of CPU cycles"


That is the basic design-requirement of all the programs I am going to
write this year ( both in C and C++ ).


I got Aho, Hopcraft and Ullman's Data Structures and Algorithms. I checked
Sedgwick too, which was good but was much expansive. Form
comp.programming I got these design ideas for an efficient program:

1.) Use a Hash-Table and then implement an external sorting routine for
Hash-Table.

2.) Use an AVL tree or Red-Black tree, that way I do not need a sorting
routine.

3.) Is uour design ( creating a new Binary tree form an existing
Binary-tree) okay from this context of efficiency ?

Though I heard about these Data-Structures, I never ever used them or
learned them. I am learning about them now.



--
http://lispmachine.wordpress.com/
my email ID is @ the above blog.
just check the "About Myself" page