Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lukego
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
37 ms
·
361.
▲
by
lukego
12y ago
Is the Enhanced Networking on a C4 instance a Virtual Function (ixgbev driver) or a full Physical Function (ixgbe driver)? The docs say that it is a VF but I wonder since they are offering the full 10G of bandwidth. We could relocate our Sn
362.
▲
by
lukego
12y ago
Sure: Intel optimization reference manual: http://www.intel.com/content/dam/www/public/us/en/documents/... Agner Fog's Software optimization resources: http://www.agner.org
363.
▲
by
lukego
12y ago
Hacking for the latest and greatest Xeon CPUs feels a lot like Forth on small machines to me. Each core has only 256KB of fast private memory (L2 cache). Squeezing the working set into that memory means fast execution that is independent of
364.
▲
by
lukego
12y ago
Isn't it amazing how well Emacs development scaled without basic features like namespaces and lexical scope? These days I find myself more inclined to try and borrow ideas from Emacs Lisp than to retrofit modern ideas back onto it. How
365.
▲
by
lukego
12y ago
Could you map all such buggy devices on the internet and treat their known addresses specially? (Scan the whole public IPv4 range every month to find them?) Just an idea :-)
366.
▲
by
lukego
12y ago
The best way to contribute to LuaJIT seems to be to learn its internals and make contributions, or to fund somebody else to do so. Mike encouraged this when he gave a talk on LuaJIT in Zurich last year. His own commercial ambitions seem to
367.
▲
by
lukego
12y ago
Hillis & Steele also wrote up many of the same ideas in a non-Lispy way. "Data Parallel Algorithms" http://cva.stanford.edu/classes/cs99s/papers/hillis-steele-d...
368.
▲
by
lukego
12y ago
Cool :). I also learned a lot about networking by writing a TCP/IP stack in Common Lisp. http://lukego.livejournal.com/4993.html if you are interested.
369.
▲
by
lukego
12y ago
I mean, did the median founders have a good experience overall? What is the distribution of personal satisfaction amongst founders? (Is that a power law too where nearly everybody is miserable and wishes they stayed in school, but the billi
370.
▲
by
lukego
12y ago
I would be really curious to know what the median outcome is for YC companies and their founders.
371.
▲
by
lukego
12y ago
loadgen is a good first example program. That is a load generator that can transmit an arbitrarily large amount of traffic (hundreds of Gbps) from a trace file using a negligible amount of CPU. This is practical and being used every day. h
372.
▲
by
lukego
12y ago
Snabb NFV is the main software being built on the Snabb Switch base right now: http://snabb.co/nfv.html
373.
▲
DynASM with Lua mode
(luapower.com)
97 points
by
lukego
12y ago
|
2 comments
374.
▲
by
lukego
12y ago
Thanks for the pointer! Do you also know a lightweight way to compile pcap-filters into BPF code?
375.
▲
by
lukego
12y ago
Tangentially: I am interested in sponsoring an open source (Apache License) implementation of BPF bytecode in DynASM (or LuaJIT). The use case is to execute packet filters compiled by libpcap. Ideally we'd win shootouts (like http:&#x
376.
▲
by
lukego
12y ago
> Why do you do it? Not everybody does. You sound like you might enjoy Mustachianism. That's the idea of keeping a lifestyle much like yours and using the low costs to work less, even while "growing up" and raising a famil
377.
▲
by
lukego
13y ago
Many people just moved on over time for non-monetary reasons.
378.
▲
by
lukego
13y ago
Cool :-). I worked at a very successful company that used a similar model: flat 50% of profits paid to employees as bonus. This worked fantastically well in the short term. Now, 10 years later, the company is still very successful but the u
379.
▲
by
lukego
13y ago
Cool. I found the hash maps and jemalloc binding on Github via your profile. That's something I imagine we will need in Snabb Switch eventually. Now we know where to find it! Thanks.
380.
▲
by
lukego
13y ago
The OLPC XO is the best machine I've had since my Amiga 500. I spent many pleasant afternoons lying in the sun hacking Forth on it. Fond memories. http://lukego.livejournal.com/8760.html http://lukego.livejo
381.
▲
by
lukego
13y ago
Slip on sunglasses, slop on sunscreen, and slap on a hat!
382.
▲
by
lukego
13y ago
That sounds pretty reasonable to me. If you can express how you want to filter with a fancy pcap-filter expression the tcpdump is the easy answer. Otherwise you might want to code it up in Lua with snabbswitch. Here is our basic trace store
383.
▲
by
lukego
13y ago
No. Honestly, software defined networking is the idea of replacing all of your networking staff with a very clever distributed Java program. Half the world thinks this is genius and the other half thinks it's a facepalm. Network Functi
384.
▲
by
lukego
13y ago
True. Or we might end up sharing drivers between projects too. In my experience high-end systems choose hardware to match software, so you only need to support one suitable option. The main reason we would add support for a new NIC is if it
385.
▲
by
lukego
13y ago
I'm the Snabb Switch originator. The project is new: I and other open source contributors are currently under contract to build a Network Functions Virtualization platform for Deutsche Telekom's TeraStream project [1] [2]. This is
386.
▲
by
lukego
13y ago
Fantastic :-). This is also my way of thinking and the way we are doing things in Snabb Switch. I'm influenced by my brief time working with Mitch Bradley on firmware hacking at OLPC. Linux would never get the same performance out of h
387.
▲
by
lukego
13y ago
That's how I write LuaJIT code for networking and device drivers too. Very neat.
388.
▲
by
lukego
13y ago
The police here in the Zurich area can be pretty flexible with car safety. This guy got a permit to install a wood stove in his Volvo: http://metro.co.uk/2012/02/13/swiss-man-replaces-cars-broken...
389.
▲
by
lukego
13y ago
Packets between internet hosts spend about 10,000,000ns in transit and they traverse around $10,000,000 worth of network elements on the path. These numbers are both worth optimizing. I'd say that the lowest hanging fruit for operating
390.
▲
by
lukego
13y ago
> For an operating system latency is more important than through put One interesting exception to this rule is operating systems for network packet routers. These need to be optimized for throughput over latency. I mean, the difference b
More ›