Ubuntu Linux Performance Over the Past Six Years on an Intel Xeon Server | Hacker News Reader

Ubuntu Linux Performance Over the Past Six Years on an Intel Xeon Server

phoronix.com

55 pointsc487bd6215 comments

Impossible to look at on mobile. The charts run off the right side of the screen. Viewing in desktop mode doesn’t seem to work. I’m guessing they’re trying to use a responsive layout, but they didn’t test articles like this on mobile. Come on Phoronix.

I read and viewed everything just fine on my Android. Maybe user error.

The right side of the images are cropped for me on iOS

Confirmed on Chrome on iOS

On Android FF it's possible to side-scroll the images at least, but yeah, not on new (iPhone 8) or older (iPhone 4s) iOS+Safari.

Surprised by the decreased socket performance, are sockets not a first-class citizen anymore? Everyone just went to http?

Anyone have clues as to why this has degraded so much over the years?

Do you know if this benchmark refers Unix sockets or also to network sockets?

>The socket activity was hit hard by Spectre/Meltdown patches

Did you even read the article?

> Please don't insinuate that someone hasn't read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that."

https://news.ycombinator.com/newsguidelines.html

I always assumed these guidelines to be more of a joke, as they also say

>Please don't use Hacker News primarily for political or ideological battle.

which HN staff themselves blatantly violate.

So because others sometimes violate the rules, it's okay for you to ignore the rules entirely? I'll leave it as an exercise to the readers to bring that principle to its absurd logical conclusions.

The Romans had a saying for this: "Quod licet Iovi, non licet bovi". While 20 centuries might have gone since it was coined it still holds true.

Going back to the original assumption that someone did not read the article it should be noted that socket performance in 18.10 is lower than that in 12.10 even when the spectre/meltdown mitigation is deactivated.

Wow how come the results are all over the place? I would've expected a 5% - 10% improvement here or there, but look at this!

https://www.phoronix.com/scan.php?page=article&item=ubuntu12...

What is exactly is being measured in, for example, the R script? The same code running in the same modern version of R on Ubuntu 12 and 18? Or is the R version dating to Ubuntu 12? Or is it contemporaneous for both releases?

I really wish these websites would repeat these 100x or something so we could get an idea of the variability in the measurements.

The versions used are listed in the benchmark image for almost all of the tests run. Here they are for R.

    Ubuntu 12: R v2.15.1 (2012-06-22)
    Ubuntu 16: R v3.5.1  (2018-07-02)

Reply on news.ycombinator.com