Tuning ShenandoahGC with 420 GB heaps

[Jeroen Borgers]

Haha, nice :-)

Op di 5 mrt. 2019 15:38 schreef Aleksey Shipilev <shade at redhat.com>:

> On 3/5/19 2:49 PM, Jeroen Borgers wrote:
> > Right, that makes sense!
> >
> > Would maybe be helpful to add some explanation on this to the text,
> maybe something like 'Pacing
> > delay includes STW time of pauses that hit while threads are being
> paced. For this reason you may
> > see delays longer than max pacing delay.'
>
> This would be enough?
>
>   Pacing delays are measured from entering the pacing code till exiting
> it. Therefore,
>   observed pacing delays may be higher than the threshold when paced
> thread spent more
>   time in the pacing code. It usually happens when thread is de-scheduled
> while paced,
>   OS takes longer to unblock the thread, or JVM experiences the STW pause.
>
> -Aleksey
>
>