Feedback request: priority queues in containers

[Johan Tibell]

I really recommend using Criterion to get rid of some of the many
problems associated with benchmarking.

http://hackage.haskell.org/package/criterion

Cheers,
Johan

On Wed, Mar 17, 2010 at 9:35 AM, Jim Apple
<jbapple+haskell-lib at gmail.com> wrote:
> I should also note that these timings are quite temperamental on my
> computer; using profiling or removing the performGC calls can make C
> faster than D and B faster than A.
>
>> Binomial priority queue timings (in milliseconds, averaged over 20 runs)
>> A: Simple Brodal-Okasaki list implementation:
>> 199.3697
>> B: Wasserman nested type implementation, simple functions:
>> 235.0143
>> C: Wasserman's 'Quick' list implementation:
>> 179.7227
>> D: Wasserman's proposed nested type implementation:
>> 163.47515
>>
>> D is still the fastest. However, A is a good bit faster than B. Is it
>> possible that C would show a similar advantage over D if it were
>> optimized as carefully as D?
> _______________________________________________
> Libraries mailing list
> Libraries at haskell.org
> http://www.haskell.org/mailman/listinfo/libraries
>