Re: [Caml-list] kaplan-okasaki-tarjan deque (was "looping recursion")

[james woodyatt <jhw@wetware.com>]

On 30 Jul 2004, at 14:20, brogoff wrote:
>
> BTW, what did your comparisons tell you?

It told me that my implementation was enough faster that it was worth 
the price of having to convert between deques and lists in order to 
marshall them to disk or on the wire, with noticeable savings even 
after that.  I can't remember the numbers right now.

At one time, I thought it would be worth the effort of writing an 
academic paper on the subject, but I never learned how to write an 
academic paper so it will get published.  So I haven't tried.  Sigh.

One of the questions that remains unanswered is how much time my 
implementation of Kaplan-Okasaki-Tarjan wastes being inefficient about 
handling the fixed size buffers.  My gut tells me there's room for 
improvement there, but not enough to overcome the advantage my 
implementation has over it.  I need to measure that.


― ∴

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners