81–90 of 129 posts
I have a bad habit of checking hackers news to know when my wifi is up or down.. I'm more productive when wifi is down because I'm not tempted to read articles. Saturday will be productivity day!
Earlier quoted context omitted.
And yet, here you are, on a site run off a flat file database.
A commenting site that: - Has average latency over 500ms when not under load - Performs quite poorly under load (I hate to bring it up, but the most recent example was Aaron Swartz's passing. Anyone who used HN then to get news knows how poorly HN performs under load) - Is restarted every week or two because it leaks memory - Keeps XSRF tokens in memory and loses them across restarts - Doesn't have a full markup lang…
With regards to the commenting site itself, I can think of no more viscerally enjoyable a forum I've ever participated in, with the possible exception of *Forum on MTS. There is nothing whatsoever that I would change about it, with the one possible exception of tweaking the markup so you could add fixed-width text/lists that wrapped over multiple lines. It's the only additional feature I've ever wanted out of HN. There is beauty in it's simplicity. [Edit: Okay, I would also move the upvote/downvote arrows a bit for mobile usage. It's almost impossible to hit the right one without a lot of zooming]
And, with rare exception of a MSM hit, the performance is more than adequate for an environ that should be encouraging reading, digesting, and composing.
I have a bad habit of checking hackers news to know when my wifi is up or down.. I'm more productive when wifi is down because I'm not tempted to read articles. Saturday will be productivity day!
profile, procrast settings use them! they are amazing!
Given that you've got performance issues and a fairly limiting deployment model, I never understood why you didn't get the most absurdly overpowered machine possible. (I assume you're not, because if you were, you'd be upgrading every ~6mo or so as faster single-core machines come out)
Would the CPU really be the bottleneck for HN?
If I were specing a machine for HN, naively, it would be a competition between a Xeon "enterprise" CPU with huge cache and memory bandwidth (interleaved up) and a gaming/desktop CPU with maxed-out single-core performance. Xeons can do single-core turboboost now, so E5-2690 which goes up to 3.8GHz is probably the best bet, but a desktop i7-3970X 4.0GHz might be an option if you don't need ECC (which also gets you slight speed improvement on memory).
Do you think you could go down for a week? That's generally how long I need to break an addiction
Add me to the list - I'm in for $100 if we can make it a one week outage. $250 if we can stretch it out to two weeks. I'm sorry, but 10 minutes on a saturday morning does nothing for me.
Can you please post a picture of the server so we can see all the sweet LEDs on network cards?
http://imgur.com/VuSH2j5
http://ed-thelen.org/comp-hist/vs-bendix-g15.jpg
[one of the first computer models I ever wrote for - a three address machine: first a source operand, second destination, third address of next instruction. The memory was a rotating drum]
Does all of HN run on one single server?
I think a lot of people underestimate how powerful a single well configured server can be. You don't need Heroku/AWS for everything.
