What's the status of the garbage collector?

Hey guys, Interesting work on the new engine... I read your papers and am concerned about the 4000% storage/memory penalty that is paid for an append-only system. I am currently using MySQL and thinking about RethinkDB after having switched over to SSD drives. Here are a few questions: - What progress have you made in the area of garbage collection within... more » By Tariq  - May 10 2010 - 2 new of 2 messages

	RethinkDB Newbie/Tutorial Questions?
	Hi All: Here's a really silly question that I'm sure will be asked in the future once your project really takes off: Would I use RethinkDB the same way that I would use MySQL? I noticed that there was a caveat when creating a new table but I didn't quite see anything else when it came to directions.... more » By Donnie Demuth  - Apr 5 2010 - 2 new of 2 messages

	Testing RethinkDB using the Random Query Generator
	Hello RethinkDB guys, I took the liberty of running a few tests using the Random Query Generator: [link] against the published version of RethinkDB and found plenty of issues that you may wish to look at. The tests that are available with the tool are described at: [link]... more » By Philip Stoev  - Apr 1 2010 - 2 new of 2 messages

	RethinkDB on smart phones
	Sounds like a great product! Smart phones could benefit most with your technology. For example, Android framework has SQLite for basic data storage and RethinkDB could be a great add-on component! By Shyam  - Feb 21 2010 - 2 new of 2 messages

	Error loading plugin
	I am trying to load the plugin in a fresh install of mysql-5.1.39 and get the following error: ERROR 1126 (HY000): Can't open shared library '/mdata/bin/mysql-5.1.39/ lib/mysql/plugin/ha_edb.so' (errno: 2 undefined symbol: my_pthread_fastmutex_init) Any idea what could be wrong? I didn't see any configure requirements... more » By naclosagc  - Oct 27 2009 - 2 new of 2 messages

Hybrid solid state and magnetic disc hybrid?

Hello, I think your product seem promising. However, in our case we would have a huge amount of data and would be interested in minimizing the need for solid state discs (cost efficiency). Besides we believe in traditional magnetic discs ability to compete with solid state discs for quite some time when it comes to sequential / continous disc... more » By gilroitto  - Aug 28 2009 - 2 new of 2 messages

	embeddable api
	do you have a version of the code that can be built as a shared library with a given set of api's so that it can be embedded into an application. I have built something myself that has high level similarities (lock-free, parallel and optimized for SSD) but I would prefer to use something like rethinkdb so that I can concentrate my... more » By jmbnyc  - Jul 29 2009 - 2 new of 2 messages

	Licensing/Pricing model
	What will be the licensing/pricing model? Will it be a paid product, or a free product with paid support? By Matt Shields  - Jul 29 2009 - 4 new of 4 messages

questions about 'Rethinking Database Storage' whitepaper

What is the cause for 'index data structures generate 2GB of data per index for every 50MB of data'? Is this from having multiple copies of data? Or from too much metadata per row? Or from something specific to your algorithm? I might disagree with your claim that it is OK to queue writers because there are few long-running transactions. Even short-running... more » By MarkCallaghan  - Jul 28 2009 - 6 new of 7 messages