At KubeCon + CloudNativeCon in London, I had the pleasure of catching up with David Nalley, Director of Open Source Strategy and Marketing at Amazon Web Services (AWS). This engaging conversation covered everything from AWS’s open source commitments to the remarkable growth of community-driven projects like Valkey and OpenSearch.

KubeCon: The Must-Attend Open Source Event

When asked about his impressions of the event, Nalley emphasized KubeCon’s significance in the open source landscape, “This event has become a must-attend. It feels like one of the largest open source events in the world. So many projects—even those not part of the Cloud Native Computing Foundation (CNCF)—use this as a nexus for coming together and uniting their communities. It’s a really powerful place.”

AWS made several important announcements at the event. Nalley highlighted their continued commitment to Kubernetes infrastructure support: “We announced earlier this week that we’re renewing our commitment to providing infrastructure for Kubernetes. It’s really expensive to build, test, and run all the performance testing required for Kubernetes. That’s why we commit up to about $3 million a year to the CNCF to support that infrastructure.”

Additionally, AWS announced a community add-on focus for Kubernetes: “We announced a community add-on focus—essentially a catalog that allows folks to take all of the CNCF add-ons and easily plug them into their EKS infrastructure. We’re providing a consistent interface, whether you’re consuming things from the AWS Marketplace or just want to use a community-developed add-on for Kubernetes.”

The Intersection of Open Source and AI

When discussing open source in the artificial intelligence (AI) landscape, Nalley thoughtfully addressed the complexities: “This is a really complicated issue, right? So much of AI is built on open source. People are using PyTorch for training, and tools like LangChain and CrewAI to build and connect models and agents. Open source really plays a foundational role in both the development and enablement of AI.”

Nalley touched on the ongoing definitional debates in open source AI: “There’s a lot of controversy around the definition of what open source AI actually is. I don’t think that question has been settled yet. Without access to the data—which very few models provide—it’s really hard to recreate or modify them.”

Nalley noted that returning to open source fundamentals helps frame these discussions: “If you look at the history of open source—going back to things like the Debian Free Software Guidelines, and before that, the Four Freedoms from the FSF—the permission to explore how something works, modify it, and distribute changes seems to be missing without access to the raw data.”

OpenSearch: From Fork to Foundation

Nalley shared updates on OpenSearch, which began as a fork of Elasticsearch after its license change: “We created a hard fork of Elasticsearch called OpenSearch, and we spent the first couple of years building the infrastructure needed to release, test and get those processes in place to create a successful project.”

The journey from AWS-led project to independent foundation was deliberate: “We spent a couple of years working to build a critical mass within the community, with the end goal of eventually moving OpenSearch to a foundation. We made good on that trajectory last year when we announced the transition of OpenSearch to the Linux Foundation.”

Valkey: A Community Success Story

Valkey, which Nalley enthusiastically described as his “favorite,” represents an extraordinary open source success story following Redis‘s license change: “The community—and I truly mean the community, including existing Redis maintainers—moved very quickly. They forked the codebase from the last version of Redis, began development almost immediately, and announced Valkey just eight days after the license change was made public.”

Nalley marveled at the unprecedented speed of this community mobilization: “If you’ve been around open source for any length of time and have seen some of these foundations form, you know that getting something done in six months is considered fast—so eight days is unheard of.”

The conversation concluded with insights into Valkey’s latest release: “Version 8.1 was just announced this morning. The most interesting feature—it’s AI-related—is vector search. If you’re looking for a true community success story, this is one, within the broader success story that is Valkey.”

Nalley highlighted the remarkable collaboration between competitors: “Both Google and Amazon had independently created vector search implementations. They began collaborating and ultimately decided to go with Google’s implementation.”

Beyond new features, the performance improvements are notable: “Version 8.0 saw a 20% performance improvement in memory usage. Version 8.1 is seeing a huge boost as well, including a 20% performance gain on encrypted data streams.”

The Individuals Behind the Projects

While companies like AWS, Google, and Oracle support these projects, Nalley made a point to recognize the individual contributors driving Valkey forward: “There are a lot of rock stars—Madeline Olsen, Ping Xiu, Bin Win, Yi, and Viktor Söderqvist—who are the current maintainers. These folks are all rock stars. They’re the ones keeping this community-driven project alive and leading it through a pretty tumultuous time.”

As our conversation wrapped up, Nalley reiterated AWS’s continued commitment to open source: “Open source remains super important to AWS and our customers, and we’re going to continue investing in it.”

The vibrant open source ecosystem, from community-driven projects like Valkey to foundation-backed initiatives like OpenSearch, continues to thrive with support from both individual contributors and major cloud providers like AWS.

Guest: David Nalley (LinkedIn)
Company: AWS
Show: KubeStruck