Java 8 curiousity

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Java 8 curiousity

Rob Tompkins
Hey all,

At my day job we saw a 60% performance improvement (cpu utilization) between 1.8.0_40 and 1.8.0_121, and I was wondering if anyone else out there has seen anything like that before or if anyone might know what could cause that given that the release notes  don’t directly point to anything.

Cheers,
-Rob
---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Java 8 curiousity

ole ersoy
Are you using streams?  A while back I experimented with matrix multiplication using streams and they can be very slow:

https://stackoverflow.com/questions/35037893/java-8-stream-matrix-multiplication-10x-slower-than-for-loop

So I would think that Java has a lot of performance tuning and optimization it could do in this space.

Cheers,

Ole


On 10/09/2017 04:20 PM, Rob Tompkins wrote:

> Hey all,
>
> At my day job we saw a 60% performance improvement (cpu utilization) between 1.8.0_40 and 1.8.0_121, and I was wondering if anyone else out there has seen anything like that before or if anyone might know what could cause that given that the release notes  don’t directly point to anything.
>
> Cheers,
> -Rob
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Java 8 curiousity

John Patrick
i've experience the same, a build about ~75 and another around ~95
introduced from the limited versions I've tested introduced some
dramatic performance and memory improvements

i think one was around streams and the other around lambdas

but that is also about u40 was April 14, 2015, and u121 was April 18,
2017, so you have 2 years of people using java 8 with real life
applications and feeding back those changes

John


On 10 October 2017 at 00:03, Ole Ersoy <[hidden email]> wrote:

> Are you using streams?  A while back I experimented with matrix
> multiplication using streams and they can be very slow:
>
> https://stackoverflow.com/questions/35037893/java-8-stream-matrix-multiplication-10x-slower-than-for-loop
>
> So I would think that Java has a lot of performance tuning and optimization
> it could do in this space.
>
> Cheers,
>
> Ole
>
>
>
> On 10/09/2017 04:20 PM, Rob Tompkins wrote:
>>
>> Hey all,
>>
>> At my day job we saw a 60% performance improvement (cpu utilization)
>> between 1.8.0_40 and 1.8.0_121, and I was wondering if anyone else out there
>> has seen anything like that before or if anyone might know what could cause
>> that given that the release notes  don’t directly point to anything.
>>
>> Cheers,
>> -Rob
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [hidden email]
>> For additional commands, e-mail: [hidden email]
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Java 8 curiousity

Simon Spero
In reply to this post by Rob Tompkins
Could be gc, could be hotspot compiler, or could be library

There was a massive improvement in the performance of the G1 GC in u60 when
dealing with "humongous"  objects; however I assume you wouldn't have been
using G1 before u60 for anything real (that was when it first became
usable.

If you attach visualvm and sample, you should get a better idea of were the
performance wins are coming from.

wrapping the app as a pseudo jmh benchmark and running using perfasm as the
profiler (needs hsdis, Linux, and perf) may be more precise (e.g. better
inlining, different instruction use, etc).

Simon

On Oct 9, 2017 5:21 PM, "Rob Tompkins" <[hidden email]> wrote:

Hey all,

At my day job we saw a 60% performance improvement (cpu utilization)
between 1.8.0_40 and 1.8.0_121, and I was wondering if anyone else out
there has seen anything like that before or if anyone might know what could
cause that given that the release notes  don’t directly point to anything.

Cheers,
-Rob
---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]