blog
SREday London 2024: Our conference highlights
From Conf42 to SREdays, the brothers Mark and Miko Pawlikowski bring developers, DevOps engineers, analysts, executives, etc from all over the world to learn from each other. Conf42 is a series of online tech events that happen throughout the year, while SREday is an in-person event in major cities worldwide.
In June, Severalnines spoke at Conf42 Observability on how ClusterControl can help organizations simplify database troubleshooting. In September we gave an in person presentation at the SREday London.
This post will summarize our presentation, and share highlights from our experience – from mind-blowing talks to eureka moments.
Database Operations (Ops) Your Way
Databases are the foundation of software applications. They serve as the organized repository for data, enabling applications to store, retrieve, and manipulate information efficiently. We all know there are different types of data models, different types of databases, and different ways to host them.
Most conversations around databases today have gone beyond deciding what type of database to use, to “how best to deploy the database for scalability and cost efficiency”. And a recurring debate is whether to deploy on Kubernetes or stick with VMs and bare metal.
At SREday London 2024, we presented on how cloud native technologies are the building blocks of modern database software, showing how organizations are increasingly deploying and scaling their databases on Kubernetes with the help of Operators and with Kubernetes using platforms such as CCX.
During the event, some attendees we chatted with were excited about databases on Kubernetes, while some shared that they wouldn’t let Kubernetes near their databases.
The choice to deploy on Kubernetes, with Kubernetes, or any other means depends on your needs and capacity. What is more important is that you are doing your “Database Operations Your Way”, hence the importance of Sovereign DBaaS.
A Sovereign DBaaS model empowers you to tailor your database operations to your exact needs. Choose the databases, tools, and infrastructure that best suit your requirements, whether you prefer open-source, proprietary, or a combination.
The level of control is entirely up to you. From coding operations from scratch to leveraging off-the-shelf solutions, and from on-premises to hybrid cloud deployments, the possibilities are endless. The only question is, how much control do you value and require for your data stack?
Kubernetes at the center stage in SRE and DevOps
As you’ve already noticed in the past few years, almost everyone is either going “Cloud Native” or talking about it. At SREday London, there was a good number of talks on Kubernetes. The following is brief summary of some of those talks.
Platform engineering and Kubernetes maturity
The K8s talks began by discussing the challenges and opportunities of maturing platform engineering initiatives using Kubernetes. Speakers emphasized the importance of balancing people, process, policy, and technology to achieve a sustainable path forward. The CNCF platform maturity model was introduced to guide organizations through this process.
Scaling, optimizing, and upgrading Kubernetes
Several talks focused on leveraging Kubernetes to improve infrastructure efficiency and scalability. Karpenter, a tool for managing node provisioning, was highlighted as a powerful solution for automating these tasks.
In optimizing Kubernetes cluster resources, the approach “Virtual Kubernetes cluster” was discussed. Virtual clusters are fully functional Kubernetes clusters nested inside a physical host cluster providing better isolation and flexibility to support multi-tenancy.
Additionally, speakers explored strategies for deploying and managing microservices across multiple Kubernetes environments.
Debugging and Troubleshooting Kubernetes
The conference also addressed the challenges of debugging and troubleshooting Kubernetes clusters. Speakers shared techniques for diagnosing and resolving common issues, such as memory leaks and network problems. eBPF tools like Cilium were presented as valuable aids for troubleshooting complex Kubernetes environments.
Will you promote AI to a senior engineer?
The first Keynote discussed automated incident management, where the speaker shared their experience adopting AI in their incident management workflow. The overall discussion can be summarized as, “AI is a good junior engineer.”
The distinction between a junior and senior engineer primarily lies in their level of experience, knowledge, and responsibilities. Promotion from junior to senior engineer requires a combination of factors. Experience is crucial, as consistent and successful performance over several years is often expected.
With the advancement of AI, and the creation of new AI tooling, both assisted and independent (autonomous), would you ever give AI full control? As of today…
…most likely not!!
Other AI and automation talks discussed where you would use AI more, and one is monitoring and observability. With AI’s ability to analyze vast amounts of data from monitoring tools, it can identify anomalies, predict failures, and provide proactive recommendations. This helps teams stay ahead of potential issues and optimize system performance.
Automated reporting and alerts can sometimes lead to “noise” and “alert fatigue” which results in decreased efficiency and increased stress among engineers. Another talk at the event delved into ways to group alerts based on content similarity and demonstrated methods to filter events based on various conditions.
Tackling industry challenges: teamwork, discipline, and strategy
Success in the technology space often hinges on the ability to navigate and overcome many challenges. Whether it’s facing economic downturns, technological disruptions, or intense competition, organizations and individuals alike must possess a strategic mindset and a collaborative approach to thrive.
A soothing talk at SREday London by Steve Wade who shared his experience going from professional football to technology due to a life changing injury shined a bright light on the mindset we should have today. As things get tough, it should not put us down, but cause us to innovate and grow.
Catch up on past SREday events
SREday London 2024 was a great event, hosting over 200 attendees and 50 speakers who delivered insightful talks on different areas of infrastructure engineering today. One of the highlights we got from an attendee was how great the venue was in enabling networking and collaboration.
To watch a replay of the talks I highlighted in this blog post and past talks from other SREday events, head over to the SREday YouTube channel.
At Severalnines, for over a decade, we have powered hundreds of companies to take control of their data through our products, i.e. ClusterControl and CCX.
Check out our Sovereign DBaaS Decoded podcast, where we interview industry leaders about implementing open-source databases, creating cloud vendor-neutral environments, and mitigating business risks, including data sovereignty and compliance requirements.
You can also join our “Sovereign DBaaS Community.” Whether you’re a seasoned database operations pro or a novice looking to connect, learn, and grow – join now!
Follow us on LinkedIn and X for more great content in the coming weeks. Stay tuned!