Top all-time · #15917
Luke Gorrie
Personal blog · lukego.github.iosystemsprogramming
Luke Gorrie's personal blog at lukego.github.io covers systems and programming. Since 2014, 5 different Hacker News readers have linked it 7 times. The last post was in 2013, but the archive is still widely read.
Too few sampled comments (7) to say how often HN recommends it.?
Why HN links it
Luke Gorrie's blog covers high-performance networking and software design, including kernel-bypass networking, the Snabb Switch project, LuaJIT, and firmware versus software. Commenters link it as background on these topics.
Summarised by Claude from a sample of 7 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Just for fun, here is some simple Snabb Switch example code to retransmit every packet that is received on a set of Ethernet ports. It’s part …
Most-linked posts?
- /blog/2013/01/04/kernel-bypass-networking/2 people · 2 mentions · first linked 2014-02-17
What Hacker News says?
We do know some prototype can boot. We don't know about operating system being in use in any environment for any purpose. Damn, we don't know could the browser engine written in Rust work or not, despite the fact writing a browser engine was the goal of Rust from the day one. By the way, network card driver can be written in LuaJIT (and perform well) too[1]. Shall we do it? [1] http://lukego.github.io/blog/2013/01/0…
Comment linking /blog/2013/01/03/snabb-switchs-luajit-ethernet-device-driver/
on “Thor – A minimalistic operating system in assembly and C++”This technique usually called 'kernel bypass'. http://highscalability.com/blog/2013/5/13/the-secret-to-10-m... and http://lukego.github.io/blog/2013/01/04/kernel-bypass-networ...
Comment linking /blog/2013/01/04/kernel-bypass-networking/
on “Let's Build a Modern Hadoop”Curious aside -- is there any blog / article out there where someone has tried to max out network bandwidth on an x86 box? I checked a POWER8 we have today and it has 14 PCIe3 slots (mix of 8x/16x), so theoretically one could load up 14 40GbE NICs on it. I'd be interested to see some kind of I/O shootout to see where both systems hit a wall. e.g. Snabb Switch 20x10GbE on x86 https://lukego.github.io/blog/2013/06/23/…
Comment linking /blog/2013/06/23/echoing-packets-with-snabb-switch/
on “Erlang and IBM Power8 in the cloud: super-high single-system parallelism”> You certainly know building and supporting that would cost more or less the same as building and operating a sizeable datacenter. If it succeeds. There are plenty of solutions for that already. For the simplest case of the user-space networking, you can pick a number of "off the shelf" solutions for it: http://lukego.github.io/blog/2013/01/04/kernel-bypass-networ... http://www.openonload.org/
Comment linking /blog/2013/01/04/kernel-bypass-networking/
on “The C10K problem”Or maybe even not completely GC-free [1]. What might be especially helpful is a good JIT that could reoptimize the code on-the-fly, when data patterns changes. Maybe performance level of 'data-guided optimization' provided by (controllable) GC and state-of-the-art JIT could beat down traditional approach someday. http://lukego.github.io/blog/2013/01/03/snabb-switchs-luajit...
Comment linking /blog/2013/01/03/snabb-switchs-luajit-ethernet-device-driver/
on “Systems Programming in C# by Joe Duffy”Oh thanks for the kind words :-) Relevant blog entry: http://lukego.github.io/blog/2012/10/28/firmware-vs-software... I worked with Mitch on "bringup" of the OLPC 1.5 hardware and got to see him doing amazing things. The best was when he needed to initialize the memory controller to make RAM available and so he flashed a small Forth system that could run strictly within cache and then connected to that with a serial…
Comment linking /blog/2012/10/28/firmware-vs-software/
on “Thinking Forth (1984)”I am impressed with Rust in many ways. I briefly considered Rust when shopping for the right language for Snabb Switch [1]. I decided against it in part because the community did a good job of setting my expectations. "They" told me that Rust will especially appeal to hardcore C++ hackers, which is a very different profile to our target group of casual programmers. So I was able to move on quickly from a choice that…
Comment linking /blog/2012/09/25/lukes-highly-opinionated-programming-language-roundup-2012/
on “On Rust Hate-Writing”