[Ext] Re: Why I am getting weird benchmark results between 0.99 and 100 percentile

[Vijay Lakshminarayanan]

There's a 20k increase between your 0.999 and 0.9999 but only a 30x
between 0.9999 and max.  imo, the former is odder than the last tail.
But only you can tell if that's normal or its causes.  JMH's warning
about numbers and data always applies.

Thanks
Vijay

On Thu, May 6, 2021 at 1:57 PM Sarath Babu <sarath.chalasani46 at gmail.com> wrote:
>
> Aleksey, Thanks for your quick response.
> Weird means the score values are having huge differences between 0.9999 and
> 1.00 percentile, could you please help me to understand the reasons behind.
>
> Regards,
> Sarathbabu C
>
> On Thu, May 6, 2021 at 11:02 AM Aleksey Shipilev <shade at redhat.com> wrote:
>
> > On 5/6/21 7:18 AM, Sarath Babu wrote:
> > > When I run the JMH benchmark results using simpletime mode getting weird
> > > benchmark results between 0.99 and 100 percentile, could you please
> > provide
> > > me a reasons behind.
> >
> > Define "weird"?
> >
> > > *Results* :
> > > Benchmark                                                            Mode
> > > CacheTest.measureCacheGet:measureCacheGet·p0.999                   sample
> > >               0.006          ms/op
> > > CacheTest.measureCacheGet:measureCacheGet·p0.9999                  sample
> > >             136.839          ms/op
> > > CacheTest.measureCacheGet:measureCacheGet·p1.00                    sample
> > >            4487.905          ms/op
> >
> > These look pretty normal "long tail" values to me.
> >
> > --
> > Thanks,
> > -Aleksey
> >
> >