Linux Performance

Built around several freely available benchmarks for Linux, Linux-Bench is a project spearheaded by Patrick at ServeTheHome to streamline about a dozen of these tests in a single neat package run via a set of three commands using an Ubuntu 11.04 LiveCD. These tests include fluid dynamics used by NASA, ray-tracing, OpenSSL, molecular modeling, and a scalable data structure server for web deployments. We run Linux-Bench and have chosen to report a select few of the tests that rely on CPU and DRAM speed.

C-Ray: link

C-Ray is a simple ray-tracing program that focuses almost exclusively on processor performance rather than DRAM access. The test in Linux-Bench renders a heavy complex scene offering a large scalable scenario.

Linux-Bench c-ray 1.1 (Hard)

C-Ray doesn't care much for the overclock, indicating that the bottleneck is elsewhere.

NAMD, Scalable Molecular Dynamics: link

Developed by the Theoretical and Computational Biophysics Group at the University of Illinois at Urbana-Champaign, NAMD is a set of parallel molecular dynamics codes for extreme parallelization up to and beyond 200,000 cores. The reference paper detailing NAMD has over 4000 citations, and our testing runs a small simulation where the calculation steps per unit time is the output vector.

Linux-Bench NAMD Molecular Dynamics

The Molecular Dynamics module of the test certainly prefers more physical cores , with the overclock giving the result a small raise but still lagging behind the Core i5 parts.

NPB, Fluid Dynamics: link

Aside from LINPACK, there are many other ways to benchmark supercomputers in terms of how effective they are for various types of mathematical processes. The NAS Parallel Benchmarks (NPB) are a set of small programs originally designed for NASA to test their supercomputers in terms of fluid dynamics simulations, useful for airflow reactions and design.

Linux-Bench NPB Fluid Dynamics

Fluid Dynamics appreciates the overclock, and we sit in the middle of the Core i5 parts and well above the previous generation Core i7s.

Redis: link

Many of the online applications rely on key-value caches and data structure servers to operate. Redis is an open-source, scalable web technology with a strong developer base, but also relies heavily on memory bandwidth as well as CPU performance.

Linux-Bench Redis Memory-Key Store, 1x

Linux-Bench Redis Memory-Key Store, 10x

Linux-Bench Redis Memory-Key Store, 100x

With Redis single thread speed as well as IPC is king, so with an overclocked Skylake it does rather well.

Core i3-6100TE Professional Performance: Windows Core i3-6100TE Gaming Performance: High End GTX 980/R9 290X
POST A COMMENT

62 Comments

View All Comments

  • C.C. - Thursday, March 17, 2016 - link

    First! Great Article Ian..I really wish Intel hadn't decided to stop the Mobo work around's allowing i3 overclocking.. Reply
  • ImSpartacus - Friday, March 18, 2016 - link

    Yeah, fantastic article. I loved how he ran benchmarks at various overclocks. Reply
  • edlee - Friday, March 18, 2016 - link

    This was really a shame that this article was not testing a regular i3 with a normal tdp, it would have shown a definate overclock to 4.5ghz and beating stock i5 by a good margin.

    It would become the celeron 300a of this generation
    Reply
  • ImSpartacus - Saturday, March 19, 2016 - link

    Yeah, is there a section that explains why a 6100 wasn't used?

    I admittedly still haven't read the whole article, I found the part that states that a 6100te is a very unusual oem-only part.
    Reply
  • RobATiOyP - Saturday, March 19, 2016 - link

    Hardly the 300a was a guaranteed 50% oc affecting both cpu, FSB and memory on a socket giving a clean & supported widely deployed set of frequencies, without any drawbacks. It meant a relatively cheap Celeron could compete with top of the line PII's using slower cache memory on Slot riser cards.

    The skylake BCLK oc, seems to come withdrawbacks slow downs have shown up in some benchmarks, probably due to the complexity of multiple timing domains in modern chips.
    Reply
  • cobrax5 - Monday, March 21, 2016 - link

    The awesome thing with the 300A was the 128KB of full speed cache. I beleive the PII's had double the cache but at half the speed. I loved the 300A - possibly my favorite processor of all time because of when I got it, etc. I had a friend who did the hack to go dual socket 300A's. I remember this whole problem of wanting to run 98SE, but only the NT kernel supported multiple CPU's/sockets/cores (all the same back then...memories).

    Anyone remember the Voodoo 1/2 add-in cards? Those things were pretty sweet for what they did for 3D games, despite the funny VGA passthru cable...
    Reply
  • 0ldman79 - Monday, April 04, 2016 - link

    Voodoo 2 and the 300A. The good old days.

    I didn't get to play with the 300A, but I got the Celeron 500 and 533, they'd hit 700+ if done right. I got to play with dozens of them and find a good one. It was fun overclocking a Dell.
    Reply
  • RobATiOyP - Sunday, March 20, 2016 - link

    The point is that CPUs get thermally limitted, increasing volts can increase Watts in a very small area. Therefore there's some sense in trying out a power efficient chip, which has headroom.

    What the benchmarks really seem to show, is to do well on multi-threaded you need.. 4+ cores. In single thread the cheap Pentium and this i3, do well against the more expensive stock chips.
    Reply
  • Flunk - Thursday, March 17, 2016 - link

    Interesting article, although it is a little bit skewed to compare the stock performance of that i5 6600 vs the overclocked i3 without including overclocked numbers for the i5, which you could have gotten using the same motherboard you tested the i3 on. Reply
  • Ian Cutress - Thursday, March 17, 2016 - link

    That might be in a future piece. Depending on how open base clock overclocking is going to be, at this point I'm wondering if each Skylake CPU I get in should have the overclock treatment given how so few motherboards enable it. Reply

Log in

Don't have an account? Sign up now