 Hi, Jeff Rick here. We're on the ground at Cassandra Summit 2014 in San Francisco, California at the lovely Western St. Francis. I'm joined now by Jonathan Ellis, co-founder and CTO of Datastacks, who just came off of an extremely technical keynote. Jonathan, welcome. Thank you. I was hanging with you for a while on the keynote, but then you kind of left me in the dust, so I know that's kind of the purpose, right? It's a really technical developer focus conference. It is, yeah. We have a technical audience, and I think they were coming to hear that kind of stuff. Yeah, it sounds like it. You got a few applause here and there, and they were certainly paying attention. So, talk a little bit about 2.1. That's the new release that you're excited about. Obviously, teams have been working hard on it. What are some of the major milestones that you were able to accomplish that you weren't able to do before? There's really three areas that we've made progress in 2.1. First is just performance across the board, that we've improved that by over 50% and sometimes over 100%. We've also made some improvements on our data modeling side, so we allow you to create custom types in the database now and let you model hierarchical information in a Cassandra row. So, now you can take your data model in your Java application and map that one-to-one to types in a Cassandra row, and you don't need to go through an object relational map or anything like that. So, it's a big win for people doing denormalized data in Cassandra. So, it's interesting. It's a very technical talk, right? But a lot of the applications you guys talked about and a lot of the customers you talked about are really fun consumer-focused customers, right? Sony, Spotify, etc. So, talk about some of the unique challenges that media presents in this world that you guys are able to help them overcome. Yeah, it's um, so I've, you know, my career has been an engineer's path and so, you know, I've got five kids and it's a little difficult to explain to them what does dad do? He sits in front of the computer all day, right? Which you tell them not to do, right? Get off the computer, go outside and play. Yeah, my wife and I have different philosophies on that, but it's fun for me to be able to show them and say, you know, that, you know, Netflix and Spotify and, you know, all of these companies doing streaming are building on my software on Cassandra. And so, you know, we do have a lot of, really in that domain, the challenge that they're dealing with is scale. That's really the big one. And so, you know, when Netflix released their profiles, now it's not just one profile per account, but I can have one for me, one for my wife, one for the kids. You know, that tripled their data volume, but because they were building on Cassandra, they were able to scale to deal with that. And we're, I mean, clearly the transactional scale for traditional relational databases has been around for a long time with banks and financial institutions, but talk about kind of the growth of scale in these non-transactional demands of some of these other applications and how that's really changed over the last several years. Well, I think if you're looking at the bank, for example, you know, you're doing, what, you know, two or three transactions a day, maybe. And with a lot of today's modern applications, you're doing two to three transactions a second per user. So, you've got orders of magnitude, more data velocity to deal with. And that's really where Cassandra grew out of was that internet generation of modern services that are driven by that speed of data. So, you're a smart guy. You've been working on 2.1 for obviously a little while before it went GA today. What's keeping you up at night? What's next? What's the next big mountain to climb? We've got a number of things that we're working on for 3.0, but I think a big one is going to be global indexes. So, to get a little technical, right now our indexes are done locally per machine. So, if I want to find all the users born in 1976, I need to contact all the machines in my cluster and ask them what people born in 76 do you know about? And so, what we're doing for 3.0 is we'll move all of that, all of those users for 1976 onto a single machine. So, now, even if I have a thousand machines in my cluster, I can get my answer from a single one. And so, that's a big performance increase. Excellent. Well, one last question. Talk a little bit about managing a development process when you've got your own folks inside and then you've got this really active community and how you kind of prioritize, balance, optimize for those two very different kind of inputs. Yeah, it's interesting because if you look at the next generation of databases, Cassandra's unique in that we do have a healthy community of developers that aren't funded by data stacks. So, when we did 1.0 and we did 2.0 and we'll do it again for 3.0, I sent a shirt to everyone who contributed at least one patch. And last time we did it, we did over 300 shirts. And so, these people aren't doing it for their day job. A lot of them, they're in for one or two patches and then they're like, you know what? My work here is done. It does everything I need it to do. But we've got that long tail of participation. And so, we've been very careful because I was working on Cassandra before I started data stacks. So, that open source community is kind of my first home. So, we've been very careful to make sure that we keep it, you know, we keep them as first class citizens and we do as much development as possible in the open. So, we have a kind of a wall between the open source team that I lead and then our proprietary technology teams on the data stacks enterprise side. So, Jonathan, congratulations again on the release. Congratulations on the show. A lot of great energy. I guess this is the fifth Cassandra Summit. It is. And we'll be looking forward to 2015. Do we know where that's going to be? Is it going to be back here in the western? I don't think we know yet. Oh, you don't know yet. Okay, great. Well, we'll keep an eye. Jonathan, thanks for stopping by. I'm Jeff Frick. We're on the ground. Cassandra Summit 2014, the western hotel in San Francisco, California. Thanks, Jeff.