[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: performance



Taylor R Campbell wrote:
   Date: Thu, 17 Sep 2009 22:32:47 -0400
   From: David Van Horn <dvanhorn@xxxxxxxxxxxxxxx>

There are benchmarks in the Okasaki paper. To me, the important cases are lookup and update. These things are simply not feasible using sequential lists. What this means is that algorithms where you might otherwise use a vector and mutation, you can now use a list and remain functional (and thus thread-safe, etc).

Of course.  But if these programs would otherwise use vectors, is
constant-time CONS important to them?

It's difficult to say since this was never an option for these programs, but I have frequently seen programs that convert between lists and vectors in order to enjoy the inductive structure of lists at times and the constant time access and update of vectors at others.

I'm not entirely sure what you're getting at with these questions. Are you not satisfied with the rationale in the document? Or is there some concrete change you'd like made to the specification and/or implementation?

David