Gian Merlino

@xvrl I'm wondering if you have run this code in production or a production-like environment?

> I made that change locally. I'm also building a local copy and going to try it out on a cluster — will report back with how that goes. Just...

> no, I'm hoping we can get the integration tests pass first, and I haven't had much time to figure out where things are failing. My hunch is that something...

The gist I posted was missing a `processing/pom.xml` update. That's here now: https://github.com/xvrl/druid/pull/2

Sounds quite interesting. It should be even better to manipulate the offheap memory directly, using methods like `buf.getLong`, `buf.putLong`, etc. Then there won't be any heap memory necessary, which eliminates...

@samarthjain Thanks for your comments. I totally agree about the importance of fault tolerance and good resource management. It'll be important as queries can get more complex and longer-running. >...

> Gian, Are there any algorithms that would be made possible by this architecture? (By algorithms, I mean distributed implementations of relational operators, e.g. shuffle-sort, shuffle-join, broadcast-join, hybrid-hash-join, use of...

@julianhyde Do you have a reference for what sort of algorithm you had in mind that uses such opposite-direction-flowing Bloom filters? I'd like to read more about it.

> I'm also wondering if the core of Druid is combined into one process, is it possible to make Druid a built-in component, like an embedded database system, for other...

> @gianm When you are doing a big join with filters on both sides - e.g. orders from customers in california for red products - then you only want to...