Tweaking Ubuntu 17.10 To Try To Run Like Clear Linux

[Kendji]

How practical is clear linux for daily use as a office workstation or a linux gaiming machine?

[geearf]

You might have some misconceptions about those. Arch is about being "simple" which in practice means being lazy when packaging software.

A decade ago Arch was indeed about being faster than others, while still following the KISS principle.
Somehow that got lost on the way...

@Michael: It'd be nice to have Solus in it as well, as it supposedly includes some performance tweak from Clear. Thanks for doing this test, I was really excited to see what the results were, alas nothing yet to use :/

@Kendji: Not very, as it does not support non-intel GPUs and probably lack quite some software.

[edwaleni]

Honestly these results are shocking. 20% area performance improvements from Clear Linux. It's like I didn't have to upgrade my CPU in the past 4 years.

This surely is showing up a deficiency in the Linux kernel developers's priorities if Clear Linux can whack 'em so easily and so hard. 98% of data centers use Intel processors. Surely the Linux people should be spending a lot of resource on performance?? I mean Red Hat market capitalisation is 22 _billion_ dollars!!! They'rere making money hand over fist. So I'm not sympathetic to the idea of some poor guy coding assembly in his garage. This is professional stuff.

Seems like somebody is asleep on the job here. Intel has faced some bad press recently, but hey, their devs are running circles around the Linux core devs here.

With this kind of perf delta, no sysadmin worth his salt can motivate to be running anything BUT Clear Linux.

Can you get a SME from Intel on the phone about a DB running on Clear issue at 2AM ET? How about if Clear has problem with panics at 6AM in Singapore? Can Intel provide a L3 specialist in Hyderabad at 4AM local time for a global Clear patch issue?

Performance is great and totally desirable, but has to be weighed against supportability when running in the DC. I will take a few percent less performance if it gains me more in stability.

[mikhailnov]

apt-build world will probably be the most interesting benchmrk of the last year!

[mikhailnov]

Can you try rebuilding glibc with agressive optimization?

[amehaye]

Honestly these results are shocking. 20% area performance improvements from Clear Linux. It's like I didn't have to upgrade my CPU in the past 4 years.

This surely is showing up a deficiency in the Linux kernel developers's priorities if Clear Linux can whack 'em so easily and so hard. 98% of data centers use Intel processors. Surely the Linux people should be spending a lot of resource on performance?? I mean Red Hat market capitalisation is 22 _billion_ dollars!!! They'rere making money hand over fist. So I'm not sympathetic to the idea of some poor guy coding assembly in his garage. This is professional stuff.

Seems like somebody is asleep on the job here. Intel has faced some bad press recently, but hey, their devs are running circles around the Linux core devs here.

With this kind of perf delta, no sysadmin worth his salt can motivate to be running anything BUT Clear Linux.

I think you got confused.

Read the article again, the performance delta has almost nothing to do with the kernel and everything to do with userland optimizations.

[caligula]

A decade ago Arch was indeed about being faster than others, while still following the KISS principle.
Somehow that got lost on the way...

Arch is still lighter than say Ubuntu. It doesn't start wifi and modem and firmware support drivers if you only have wired ethernet. I've never ever seen an Arch system that boots slower than stock Ubuntu.

[geearf]

Arch is still lighter than say Ubuntu. It doesn't start wifi and modem and firmware support drivers if you only have wired ethernet. I've never ever seen an Arch system that boots slower than stock Ubuntu.

What I meant is that they used to care about Arch being faster, not simply leaner, it started as 686+ when others were still on {3,4,5}86+.
Today it's just quite generic AMD64 while it could depend on newer extensions.

[vegabook]

I think you got confused.

Read the article again, the performance delta has almost nothing to do with the kernel and everything to do with userland optimizations.

I think the general idea remains valid. Why would I run anything but Clear Linux given the consistent and often huge performance delta on a wide range of end-user and server workloads?

[caligula]

I think the general idea remains valid. Why would I run anything but Clear Linux given the consistent and often huge performance delta on a wide range of end-user and server workloads?

People want support. Ubuntu provides more hand-holding, which is crucial for noobs. They don't really care if it's slower. At least it works for them.