 All right, hello, Cassandra community. I've missed you. How have you been? So here we are, so Cassandra 5. And we have a few things to announce today. Thank you, everyone, for being here and for everyone that came for AI. How about some Cassandra, too? So the big thing that we need to talk about today, let's get things started, is Cassandra 5. Where is it? When am I going to have it in my hands? So right now, Cassandra 5 beta 1 is released. It is currently released. And that's an Apache release. The next step, of course, is RC and then GA. When is that going to happen? As soon as you get out there and test it, so let's go. Let's get out there and get it done. So there are some changes that were made a couple of months ago so everyone knows what is going on. So as you know, I think if you're talking about Cassandra in the past year, the word asset transactions have probably come up a couple of times, right? Oh, man, dead crowd. What's the deal? You don't have databases anymore. Geez. All right, so here's the change that happened. So 5.1 and 5.0, what's the deal here? So part of what happened was in the project, we voted to decouple the asset transactions portions and TCM, which is a transactional cluster metadata. That's a long word, but anyway, asset transactions, we decoupled it away from 5.0 to get 5.0 out the door. So that was just a matter of just shipping code, getting things done, but also giving us a little bit of time to work with asset transactions and not have to hold up everything else. So it's a good thing, I think, for us because what we get is what we always want with Cassandra, and that's quality. Quality is the most important thing. We don't just ship code, just ship code. So 5.0 is baked, ready to go. It's going to be soon, probably in the next few weeks. 5.1 will be following shortly after. The branch is out there. It's all in open source right now, so you can do that. So if you're looking for asset transactions in my talk tomorrow, that'll be that. Another thing to bring up is, so Cassandra 5, if you didn't know this, it causes the Cassandra 3.x branch to go into EOL. We announced this last year or earlier this year, but part of that is the certifications, which are all on 3.0, the data stacks provides, will also be invalidated. So there are new certifications available as well. So you can follow the QR code. It's completely updated, includes even asset transactions as well, so nice and updated for everyone. And then finally, part of another thing to announce in the project is we now have this new program. Some of you might remember the MVP program from a long time ago. This is an updated program, completely run by the ASF project, by Cassandra Project and the PMC. And this is a Cassandra Catalyst, and these are folks that are doing a lot of really good things in the community and getting the word out and energizing the community in the right ways. And I'm happy to announce we have our first batch, and there are more to come. But if you're interested in nominating someone or even putting your hand up, we're taking nominations all the time. It's not a once a year thing. So I'm going to hand it over to Katarina. Thank you. Hello, everyone. I'm very, really excited to be here today. Thank you for joining us. My name is Ekaterina. I'm a software engineer at Datastacks and Apache Cassandra Committer since 2020. So what's in Cassandra 5 that Patrick just mentioned? A few months ago, Patrick conducted a user survey. And after that, the community delivered. So since last week, we have Cassandra 5.0 Beta. And you can see there, we already delivered the storage attached to indexing, which is tightly, the indexing mechanism is tightly coupled with the storage engine, which provides huge opportunities for the data modeling. Also, it's very performant. And it works with the existing data models, which is a great thing. Unified Compaction Strategy was also delivered. And the great thing about it is that you don't need to know anything because it just works with our already two very well-known Compaction Strategies. And it's also very performant. And it reduces the overhead from space perspective. Vector search. Have you heard of the JNAI stuff? Yeah, I guess so. I guess you heard. OK, so the community worked to deliver for you vector search, which will deal with curious with unstructured data. And that provides great opportunities, like recommendation systems, chat boards, all kind of stuff that you can imagine. Now you can do it Cassandra. But that's not all. We have also a lot of other stuff, as you can see here listed. And the good thing is that all our documentation is updated. So if you haven't been recently to Apache Cassandra website, please check it. All our documentation, log posts, everything is there. Also, another note. There are a few sessions already about most of the big features these days. So I think you will enjoy to see them if you happen to not have the time or attend another session. There is some conflict. There will be recordings. But that's not all. We keep on delivering. So what's next in the 5x? As Patrick already mentioned, first we have the transactional cluster metadata. It's already in 5.1. It's already delivered. It was a huge effort. Thank you to everyone who made it happen. And right behind it are the general purpose transactions. So who doesn't like to code less and have less time to production? But that's, again, not everything. We have also CEP29, the SQL note operator, approved as a proposal. So we hope that it will be implemented soon, too. And there are also other stuff on the mailing list. Please join us and see the discussions. Give feedback. Test Cassandra 5.1, whatever you prefer, and get anything back back to us. Thank you. Hi, Katerina. Thank you. Hi, everybody. My name is Scott Andreas. I work at Apple in our cloud services group focused on Cassandra. Patrick, Katerina, and I are here to tell you about how rapidly Apache Cassandra is advancing. I don't think there's any other database in the industry that's gaining these types of industry leading features as fast as Apache Cassandra, and none can match its scale and availability. Tomorrow at 10.30, I'll be speaking on Cassandra's evolution and where the database might be headed. This is a real quick preview. If there's one thing that you learn about Cassandra this week, I hope it's about the future of distributed asset transactions in the database and the protocol that powers them called Accord. There are a lot of ways to implement transactions in a database. You can do things like grab a consensus library and serialize transactions through them. You can nominate a single leader. You can punt to the application to implement an atomic visibility protocol. Lots of ways. We did it the hard way. Accord is a leaderless PAXOS protocol that's designed specifically for Apache Cassandra. We designed it in collaboration with academic researchers, Cassandra operators, and Cassandra users. We validated the library with a formal proof, as well as large scale simulation infrastructure. This work enables Cassandra to take a unique place in our industry. This slide's kind of an eye chart, but the bottom row is where you should really focus. Cassandra is one of the only databases that can scale to petabytes of storage in a single DB. Accord enables Cassandra to support transactions up to strict serializable isolation, the strictest isolation standard, including transactions across tables. Advance it one more time. Lots of people love that Cassandra is open source and allows them to avoid cloud provider lock-in. Cassandra's transactions are also leaderless. This means that they can scale without bottlenecking on a single leader and you can initiate them from any data center to run active-active without failover. They don't suffer from gray mode failures of a hop-up leader, and they achieve some pretty great latency properties too. As a leaderless PAXOS protocol, Accord needs only to agree on the set of dependencies in the execution order of a transaction. This enables Cassandra's transaction to execute in a single network around-trip, whether single key or multi, and from a local region or remote. But the important thing to remember is that none of this is magic, it's just PAXOS. You're gonna hear a lot about new features in Cassandra today and tomorrow. There's a lot to learn about, but here's the perspective to keep in mind as you learn about those. Today, Apache Cassandra is already a database that can serve petabytes of data at millions of QPS at 99.999% availability, which I say without exaggeration due to its dynamo roots. You can replicate Cassandra comfortably in up to five regions and probably more soon. Apache Cassandra runs in any data center or public cloud without specialized hardware, and very soon you'll be able to execute distributed asset transactions across the entirety of the database from any region. This is a really incredible set of capabilities and sets Cassandra apart within our industry. But there's one more thing that I really value the most. It's that Apache Cassandra is 100% open source under a permissive license and can be downloaded, learned from, and modified by anyone. This last part is incredibly important to me as someone whose path to engineering didn't start with a CS degree. Given an increasing trend toward proprietary databases and cloud provider consolidation, the next generation of distributed database engineers need a place to learn and a place to innovate. The Apache Cassandra project is a great place to do that. Thank you all for coming. I can't wait to see you and meet you with this conference. I hope that all of you learn wonderful things about Cassandra and AI, and that you come up and say hi later. Thank you all very much.