Hacker News Reader

zackangeloon Running an Akka cluster in an EC2 Autoscaling Group

Were you using Chef inside of OpsWorks? I'm curious about what problems you're having.

I'd think that hooking the configure lifecycle event on your cluster layer in OpsWorks would work just fine.

zackangeloon Scaling the Facebook data warehouse to 300 PB

http://aws.amazon.com/datasets

zackangeloon A Scala Corrections Library

> Is the goal solely to avoid having to explicitly state the type of the target collection?

Pretty much.

http://stackoverflow.com/questions/1722726/is-the-scala-2-8-...

zackangeloon Effective Scala

Spray

zackangeloon You are dangerously bad at cryptography

Also, this random behavior will be significantly amplified if the host is running inside of a VM.

zackangeloon Tesla Raises $1B from Stock and Bond Sale

See: Liar's Poker [1]

[1] http://www.amazon.com/Liars-Poker-Michael-Lewis/dp/039333869...

zackangeloon Introducing the Square Stand

They do.

zackangeloon All servers flooded by salt water - is it possible to recover data?

We had a flooded rack in New Orleans after Katrina. We had tape backups, but during the evacuation frenzy the wrong tape was pulled from the library (we should've just yanked the entire magazine). Not recovering the data would result in thousands of lost man-hours of work, so when we discovered that the tape we had was blank, it quickly became a do-anything-at-any-cost situation.

Things were dicey post-Katrina and they weren't letting anyone in the city. We had to charter a sea plane and land it in a canal behind our office to gain physical access. We were escorted by mercenaries armed with automatic weapons. Once inside, we grabbed every drive and tape out of every server and array enclosure.

We ended up being able to do a full recovery from tape once we got back. Sometime later, when we were able to get the entire rack over to another data center, I was curious if any of the gear would work so we reconstructed everything. Miraculously, after being submerged for over 4 days, nearly every hard drive and server booted with zero data loss (one or two drives didn't spin up, but they were RAIDed). Perhaps even more impressive is that they stayed running that way for years after in our staging and test environment. It's pretty crazy how well-engineered and reliable the server-grade stuff can be these days.

zackangeloon The Beauty of Bresenham's Algorithm

I first learned of Brensenham's line drawing algorithm from Michael Abrash in his Graphics Programming Black Book. I've yet to come across a more thorough and easy to understand discussion:

http://downloads.gamedev.net/pdf/gpbb/gpbb35.pdf