News for hackers. Stuff that matters.
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by pierrefar on Monday July 12, 2010 @04:42AM (#1507230)
Very sparse on details. Are they partitioning the data? If so how? Is it MongoDB shards or at the application level or...? What's their average record size? Do they see a long-tail kind of access such that MongoDB really does keep the most commonly used records in memory all the time? How many machines are they serving from? What are their specs (particularly RAM)?
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by kneath on Monday July 12, 2010 @04:51AM (#1507248)
This is pretty well covered in this presentation: http://www.slideshare.net/fehguy/migrating-from-mysql-to-mon...

A few unanswered questions (sharding? Looks like none to me), but a good bit of numbers and specifics on their current setup.

Re:B is for Billion: Wordnik passes 9 billion record mark wi... by ericflo on Monday July 12, 2010 @06:25AM (#1507393)
The 47.7 queries per second figure in those slides surprised me a bit. I'd like to know more details there...is that for a single node, or the entire cluster?
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by fehguy on Monday July 12, 2010 @11:09AM (#1508077)
That's per api server, we have 4 api servers.
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by meghan on Monday July 12, 2010 @09:16AM (#1507670)
Video of this talk is available here: http://blip.tv/file/3621377

This article also has some details: http://www.royans.net/arch/mongodb-migration-from-mysql-at-w...

Re:B is for Billion: Wordnik passes 9 billion record mark wi... by johnny99 on Monday July 12, 2010 @04:56AM (#1507255)
Some of these questions are answered in a separate set of slides, which the author links to in the comments of the post. Hardware looks like a single server, 2x4 core CPUs, 32 GB of RAM, and a FC SAN.
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by ericflo on Monday July 12, 2010 @06:27AM (#1507396)
Single node, eh? I thought MongoDB was not supposed to be run in that configuration.
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by thegoleffect on Monday July 12, 2010 @10:56AM (#1508024)
They're probably running 1+ Mongod process/node per core - not exactly ideal though.
Re:B is for Billion: Wordnik passes 9 billion record mark wi... by fehguy on Monday July 12, 2010 @11:11AM (#1508087)
We run one mongod master, one slave (soon to be two slaves). These are used across all the api servers.