I'm quite familiar with Velocity because I’ve been attending it on and off since 2009. When it started, the themes were about how to deploy and run large-scale infrastructure.
Since then, Velocity has moved beyond just operational topics and expanded into the development side of building large-scale applications. It’s no wonder O’Reilly decided to combine the Software Architecture and Velocity conferences, seeing how these are the most transformative topics of modern software.
“Distributed” Moves to the Main Stage
Velocity has always been about cutting-edge technologies and this year didn’t disappoint.
One of the most striking differences between this year’s conference and what I saw in 2009 is the focus on large-scale distributed applications. Building applications that scale easily and run globally is just how we do business now; it’s not optional. The increased attention toward privacy and security and all the associated regulations has made matters even more challenging in recent years. The more places we want our data, the more problems we face.
Serverless and Edge Computing—the New “Hallway Track”
Where cloud- and distributed systems-related topics have clearly moved from the “hallway track” (i.e., what attendees talk about between sessions) to the main stage, serverless and edge computing have become the new hallway track.
At DataStax Accelerate I gave a talk on data models for serverless applications that, at the time, I felt may have been a little premature. However, after talking to several developers who are beginning the journey with serverless and actively considering edge computing, I feel like this is the conversation to be having right now. Testing and development are happening in many places and it won’t be long before we see more production workloads deployed.
If you’re not familiar with serverless computing, it’s a cloud-based service where you simply supply the code and the cloud provider will run as required without any need to set up servers. This works for a variety of use cases but the most interesting one is microservices-based applications. Scaling large applications with small, composable APIs has been done to great effect in traditional infrastructure, but serverless potentially lowers the setup overhead with the same benefits.
Edge computing takes the idea of serverless one step further by distributing the runtime to closer points globally for your end users. Think of a content delivery network but for running code. Yet another acknowledgment that speed counts, and in a global economy you need to be as close to your users as possible. Of course, this discussion naturally gravitates to the most important part of your application: your data.
What to Do About Your Data
Every conversation I had about serverless and edge computing always came back around to data, and the answer I give to people asking about how to handle distributed data has always been the same: Apache Cassandra™. Cassandra was purpose-built to replicate data anywhere you need it and this hasn’t changed in the world of serverless and edge computing.
If anything, Cassandra has once again proven the unquestionable value of being able to distribute your data in a reliable and consistent way; this is the direction we’ve always been moving in and will continue to move in as the world becomes more distributed and less centralized.
And this, of course, ties into something that wasn’t new at Velocity: the incredible community members that stopped by and talked to us at our booth in the expo hall.
The Power of the Cassandra Community
The Cassandra community is huge and made its presence felt at Velocity! Many people came to talk to us about their use cases and ask for advice on new projects. Just seeing the incredible variety of companies using Cassandra for so many different use cases—from user management to customer interaction to fraud detection to messaging—reminded me of how much the ability to distribute data has changed the technology industry.
There’s a saying about Cassandra: “Everyone has a little bit of Cassandra in their environment.” Well, I can say with conviction that some organizations have a lot of Cassandra in their environment! It's the clear choice for a hybrid cloud, multi-cloud, or multi-data center deployment. It’s the choice developers are making because it’s database that was built to do what they need to have done, and it’s truly designed for the future of data management.
I plan to return to the Velocity and Software Architecture conferences next year. I skipped a few years and missed the community of users that are all about building distributed applications. I'll especially be interested to hear more about how developers are using DataStax Constellation and all the cool things that they will be building as we release more products to help them. It will also be a good time to get a check-in on adoption of serverless and edge computing within mainstream enterprises.
You can count on me being there listening and asking questions. I want to know what's happening in the world of developing large-scale applications because that’s the community that I love and live in, and it’s also the future of data management. I’m honored and excited to be a part of that future.