Yay! Hacker News is on a new and improved server now.

https://news.ycombinator.com/item?id=516108

by kirubakaran • 17 years ago

41 37 17 years ago

news.ycombinator.com

"We're down for (we hope) a few minutes while switching to a new server."

... and they are back.

Thank you pg, rtm and everyone who made this happen.

pg5• 17 years ago

Yes, should be a lot faster. We're now just trying to kill the old one cleanly...

kirubakaran• 17 years ago

It definitely feels much faster. Thanks again pg!

pg5• 17 years ago

Looks like everything went well. Handy having Robert Morris as your sysadmin. The new server seems to be about 2x as fast. The frontpage renders for me in about 50 msec. But the site should seem more than 2x faster (for logged-in users) because many requests will terminate before being interrupted.

There's now enough memory that we can fit all the links and comments in memory at once again. We should be good for another year or so.

aaronblohowiak2• 17 years ago

Thanks! Also, for those of us who don't know who Robert Morris is:

http://en.wikipedia.org/wiki/Robert_Tappan_Morris

* MIT Prof

* creator of first internet worm

* cofounder of viaweb and ycombinator

breck• 17 years ago

Are you planning on pointing the domain to the new server or just using 174.132.225.106?

It could solve the growth problem, perhaps?

pg5• 17 years ago

news.ycombinator.com already points to the new ip address. The message with that ip addr in it will only be seen by people who still have the old ip cached in their browser. (Obviously for them we have to refer to the new server by its ip addr.)

vaksel2• 17 years ago

wonder what the server specs are...for both the new and the old one

rtm2• 17 years ago

Old: 2.4 GHz Pentium 4, 4 GB RAM, 32-bit FreeBSD 5.3.

New: 3.0 GHz Core whatever, 12 GB RAM, 64-bit FreeBSD 7.1.

cperciva5• 17 years ago

You were running FreeBSD 5.3 until recently? You know it reached its EoL at the end of October 2006, right? There have been lots of security issues over the past 2.5 years which weren't fixed in FreeBSD 5.3.

And by "FreeBSD 7.1" you mean 7.1-RELEASE-p3, right? :-)

vaksel2• 17 years ago

thanks

lsb• 17 years ago

So, if you've got all those in-memory data structures, there are lots of lists of pointers to objects. How does it compare on a 32-bit OS, with 32-bit pointers, vs a 64-bit with 64-bit pointers?

rvirding• 17 years ago

Perhaps if they had been running Erlang on the server you wouldn't have had to take the site down while migrating to a new server. Sorry just had to point this out

cperciva5• 17 years ago

Erlang isn't a magic bullet. It might make migrating processes fast, but it won't speed up migrating data.

ciupicri• 17 years ago

Indeed, but I think that he was trying to say that Erlang could have made this process more transparent for the users, without any disruption.

airlabam• 17 years ago

>>Erlang isn't a magic bullet. It might make migrating processes fast, but it won't speed up migrating data.

Could you explain why to me? I just can't remember. I think I might be suffering from mnesia.

tlrobinson2• 17 years ago

So we just finished a new round of optimizations as part of our ongoing quest to prove that with sufficient caching you can serve arbitrarily large numbers of requests with arbitrarily slow languages.

http://ycombinator.com/newsnews.html#15jan09

Is that quest over, or is upgrading the server fair game? :)

pg5• 17 years ago

No, I'm sure I can squeeze out a lot more performance. But I have a lot of other stuff going on at the moment (Demo Day is this week, among other things) so it is nice to be able to buy my way out of trouble for a change.

patio11• 17 years ago

with sufficient caching you can serve arbitrarily large numbers of requests

Note the collorary: with a sufficient business model you can earn arbitrarily large amounts of money relative to hardware costs.

Which always makes me wonder why the clooooooooooooud gets so much press for cost reduction. If your application doesn't value data by weight (or worse, values data at 0), your hosting costs are going to be rounding error. Reducing the rounding error to half-sized doesn't strike me as the number one best use of time for the majority of businesses.

tlrobinson2• 17 years ago

True, though reduced hardware cost is only part of the promise of cloud computing.

Elastic scalability is a major advantages in some cases, allowing you to rapidly scale up (and back down) while minimizing waste. "The cloud" absorbs the fluctuations.

Easy deployment and thus reduced administrative costs might be another. With EC2, etc that's might not be the case, since you still have to set up the OS images and other infrastructure yourself, but things like Heroku might help.

blueben• 17 years ago

Why "server" instead of "servers"?

EliAndrewC• 17 years ago

Typically sites with multiple servers put there web and database server software on different machines. Hacker News doesn't use a database and stores everything in memory, so this wouldn't make any sense for this codebase.

Yay! Hacker News is on a new and improved server now.

Related Stories