News for hackers. Stuff that matters.

A comment on GraalVM. Parent comment

Re:GraalVM by bdibs on Sunday August 09, 2020 @11:28PM (#24105201)
And is that with their javac or is it reserved only for their native image generation?
[ Parent | on HN ]
Re:GraalVM by smabie on Monday August 10, 2020 @12:24AM (#24105547)
In my experience, native image is noticeably slower than regular openjdk, at least for Scala. The start up speed is nice, though.
[ Parent | on HN ]
Re:GraalVM by asymptotically2 on Monday August 10, 2020 @12:53AM (#24105675)
You need to pay Oracle $lots for a GraalVM enterprise licence to get profile guided optimisations, if you want some of your peak performance back.
[ Parent | on HN ]
Re:GraalVM by pjmlp on Monday August 10, 2020 @02:09AM (#24106092)
Twitter thinks otherwise, as they are the major consumer of GraalVM in production.
[ Parent | on HN ]
Re:GraalVM by didibus on Monday August 10, 2020 @02:16AM (#24106132)
GraalVM is a JIT compiler for the most part. That's the one Twitter uses I believe. SubstrateVM is what lets you perform a closed world assumption ahead of time build. And that does indeed run left performant most of the time, but it uses less memory and starts faster.
[ Parent | on HN ]
Re:GraalVM by pjmlp on Monday August 10, 2020 @03:03AM (#24106408)
Unless one feeds it with PGO data, just like it happens in most AOT workflows for optimizations like devirtualization.
[ Parent | on HN ]
Re:GraalVM by secondcoming on Monday August 10, 2020 @04:36AM (#24106913)
What part of production, do you know? Because mobile.twitter.com links are slow as hell to load
[ Parent | on HN ]
Re:GraalVM by pjmlp on Monday August 10, 2020 @04:47AM (#24106982)
Twitter has a couple of talks about it.

https://jaxenter.com/graalvm-chris-thalinger-interview-16307...

[ Parent | on HN ]