Criterium automates the JVM warmup process.

On Sunday, February 1, 2015 at 1:16:37 PM UTC-6, Ashton Kemerling wrote:
>
> Also remember to give the JVM some warm up time, as the JVM depends 
> heavily on JIT style optimizations, and measuring performance without it 
> might not represent real world behavior,
>
> --
> Ashton
>
> On Saturday, January 31, 2015 at 11:39:05 PM UTC-7, Mars0i wrote:
>>
>> You also might want to use Criterium 
>> <https://github.com/hugoduncan/criterium> rather than *time *for 
>> accurate benchmarking*.*
>>
>> On Friday, January 30, 2015 at 6:54:52 AM UTC-6, Maris wrote:
>>>
>>>
>>> yes,  it helped     :-)
>>>
>>> type hints make non-trivial difference   
>>>
>>> thank you
>>>
>>> On Friday, 30 January 2015 12:43:40 UTC, Nicola Mometto wrote:
>>>>
>>>>
>>>> If you set! *warn-on-reflection* to true, you'd see a lot of reflection 
>>>> warnings from your code. 
>>>>
>>>> Type-hinting the code like this: http://sprunge.us/ATiV makes your 
>>>> example execute in 120ms on my machine. 
>>>>
>>>> Maris writes: 
>>>>
>>>> > I implemented leftist heap (from Purely Functional Data Structures 
>>>> book) 
>>>> >  in clojure. 
>>>> > 
>>>> > https://gist.github.com/maruks/135fef92455578b61de2 
>>>> > 
>>>> > It takes 32 seconds to insert 100000 elements in heap: 
>>>> > 
>>>> > (time  (peek   (reduce conj (empty-heap)  (range 10000000 20000000 
>>>> 100) 
>>>> > ))) 
>>>> > "Elapsed time: 32649.71438 msecs" 
>>>> > 10000000 
>>>> > 
>>>> > Why is it so much slower than SML NJ?  Something wrong with my code? 
>>>> > 
>>>> > 
>>>> > Here is SML version: 
>>>> > https://gist.github.com/maruks/c863eac9cf057a071307 
>>>> > 
>>>> > And function that inserts 100000 elements in *54* milliseconds ! 
>>>> > 
>>>> > fun test(n:int) : int  = 
>>>> >     let 
>>>> > val r = List.tabulate(n, fn x => 10000000 + 100 * x) 
>>>> > val h = foldl ( fn (e, h) => IntHeap.insert(e , h) )  IntHeap.empty 
>>>>  r 
>>>> >     in 
>>>> > IntHeap.findMin(h) 
>>>> >     end 
>>>>
>>>> -- 
>>>>
>>>

-- 
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your 
first post.
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to