 So, hello, everyone. Welcome to this session on CNCF project updates. My name is Ricardo Rocha. I'm a computing engineer at CERN. And today I will give you some updates on different projects in the CNCF, some projects that have moved in their graduation stages, and also some specific updates on different projects. I will start this session by giving a quick recap on the different graduation stages that a project can go through in the CNCF. This stage is really expressed different levels of maturity for the projects, and they are really a good indication for both end users and other projects in terms of the expectations they can have. Usually the first stage for a project to join is the sandbox. This is an entry point stage. It hopefully gives the project more visibility, more exposure to end users and to other projects, and it's a very nice way for a project to grow. Next up is incubation. Incubation is already where most of the due diligence already happens, but a project also has to show very specific criteria, including at least three independent end users, a healthy number of committers, and a healthy number of commits as well. After that, a project can apply for graduation, and this is the last step in the evolution of a project. In here, a project already has to show committers from at least two organizations. It has to go through a third-party security audit. It has to show very well-established project governance and also a well-established committer process so that people can easily join the project. What I will do now is I will give you an update on projects that have moved to incubation recently. This is quite exciting because in addition to a lot of the projects we've seen just a couple of months in Cook on North America, we already see new projects also joining incubation. I will start with Cilium. Cilium is a project that secures network connectivity and load balancers between applications. Next up is Cata, which is the Kubernetes-based support event-driven autoscaler. After that, we have the distributed application Runtime Dapper, which provides APIs that simplify microservice connectivity and they do this by supporting both service-to-service invocation and PubSub messaging. This project helps writing resilient and secure microservices. After that, we have Flux. Flux is the continuous and progressive delivery solution for Kubernetes that is open and extensible and I will give some more updates about this project later. We also have Crossplane, which is a project I'm particularly excited about and it's a Kubernetes add-on that enables platform teams to assemble infrastructure for multiple vendors. They expose higher-level APIs which make for a great developer experience. After that, we have Longhorn and Longhorn is with Dapper. One of the two projects that have moved to incubation in the last couple of months is KubeCon North America. Longhorn is a distributed block storage system for Kubernetes. It is designed to run on top of different types of physical storage devices, infrastructures and architectures. Finally, we have OpenTelemetry. OpenTelemetry is a collection of tools, APIs and SDKs to instrument, generate, collect and export telemetry data so that software behavior and performance can be analyzed. With all of these movements, we now have 26 projects in incubation stage. Again, two new ones just since KubeCon North America two months ago. Thanks really to everyone for the great work and the community for the awesome collaboration that helps these projects move and reach higher levels of maturity. In addition to incubation, we have a new project that moved to graduation and that's LinkerD. LinkerD is the 16th project to join the graduated stage in the CNCF. We now have 16 projects graduated. It is the service mesh for Kubernetes. It makes running services easier and safer by giving you runtime debugging, observability, reliability and security. All of this without requiring any changes to your code. I'll move now to give you some specific updates about different projects. The first one I will talk about is OpenCelemetry. OpenCelemetry has specific goals that include providing high quality telemetry built-in in all the cloud-native software, providing a well-factored set of components which allows easy instrumentation of open source software and sending data anywhere in any format and also improving observability by integrating tracing, metrics and logs into a single stream of structured correlated data. The news in OpenCelemetry includes the GA of the tracing spec with version 1.0, now available in multiple SDKs. This includes Java, .NET, Python and Go. The beta for metrics implementation with interoperability with Prometheus being something that is key and also working towards full compatibility with OpenMetrics. The log specification work is now in beta and of course if you want to check more details visit this link here with the OpenTelemetry.io status and you can see the different status of the multiple components that make OpenTelemetry. Next up is Flux. I mentioned I would give some more news about Flux. The big thing is of course Flux V2 that has been growing in usage and that has triggered the move to incubation. There have been really significant updates to the project that needs highlighting. The first one I'll mention here is the integration into the GitOps offerings of multiple cloud vendors. The introduction of server-side-apply, so this really brings much improved CPU, memory and network performance. It also dramatically reduces the number of calls to the Kubernetes API. After that we have a set of stable APIs from now on which is really of course well received by everyone. And finally I would mention the drift detection between the desired state and the cluster state done reliably. This is great news as it means that users can now wait for all the applied resources to become ready and they don't have to write any more health checks themselves. The news also about Lingardy, big news graduation of course, the huge news, but also some specific updates on the project. The first one is authorization policy. This allows end users to easily enforce which types of connections are allowed with a cluster and this is done based on TLS identity and also benchmark results. In this project the benchmarks mean that Lingardy users can minimize the resource requirements and the user-facing latency that is imposed by running a service mesh. If you want to get involved do it by checking Lingardy.io. After this I will go on a lightning round of project updates to cover as many as possible in this small amount of time. The first one is PCV Inspire. These are in the SDKs and APIs which allow users to easily integrate and test new plugins and automation codes. The second one is in the controller. This allows user-gen consumption of PCV Inspire via pure Kubernetes API interactions so no need to understand the in-depth mechanisms that leverage the project. Much welcome to improve DZO views of course. And finally in integration with serverless platforms specifically it can now deliver cryptographic workload identities with certificates or JSON web tokens to service workloads from cloud platforms like AWS Lambda and Google Cloud Functions. GRPC is a very popular project. The big news here are the recently introduced retry and session affinities. This ability to retry helps improve service availability by enabling GRPC applications to retry outbound requests according to a policy. This can really be very helpful if instances or endpoints that host your applications are slow or flaky. After that we have Prometheus, Prometheus toolkits. Here the updates include high resolution histograms. This is a feature that should solve a lot of the errors in quantile estimation. It's a very, very cool feature. And also the launch of the Prometheus conformance program. This program enables end users to determine which projects or services are truly compatible with Prometheus and they will hopefully prevent fragmentation and interoperability and ensure improved reliability as well. VITES. VITES is a database solution for deploying, scaling and managing large clusters of open source database instances. Some updates here include the improvements in shorter latencies and reduced CPU memory and memory usage for serving queries. This means that the system can both deliver throughputs but also lower the hardware and cloud costs to render the service. And second last, we have Fluentee. Fluentee is an ecosystem to solve the logs collection and processing in containerized environments. The big update here is that this project is no longer just for logs. It has first class integration with open metrics and Prometheus and there are plans to do the same for open telemetry. So please give this a try and provide feedback to the project. And finally, Core DNS. Core DNS is a flexible DNS server that many of us run for service discovery inside Kubernetes clusters. The big update here is that the Core DNS team is moving more of the DNS functionality into plugins and this includes things like the new zone transfer plugin. The other piece of news that is quite important as well is support for it's an ACME plugin which is in development that should allow supporting automated certificate management through the ACME protocol. And with this I come to the final slide of this presentation. Again, thank you everyone for the amazing collaboration in the different projects and all the end users for supporting and helping these projects reach higher levels of maturity. And I wish you all an amazing KubeCon.