ConfPad

Paul reviews the journey of the development of InfluxDB from version 0.9 in 2015 to the 2.0 version that they are working on in 2018. He talks about building their own storage engine, moving from a monolithic database to a services based containerized data platform and scaling out the development team.

Must we really orchestrate our containers, or could they perhaps just do some improv jazz? The container landscape has converged on Kubernetes. Bridget talks about how, what, why, and other such excellent questions. You’ll come away with ideas for learning and sharing about k8s.

In a distributed data store, the CAP theorem forces a choice between strong consistency (CP) and availability and responsiveness (AP). To address this issue, we take an application-driven approach, Just-Right Consistency (JRC). JRC derives a consistency model that is sufficient to maintain the application invariants, otherwise remaining as available as possi…

How do you demonstrate your software is ready? However you define "demonstrate" or "ready", sharing how applications fits requirements is important not only internally at your organization, but possibly valuable to community peers worldwide. Julio addresses five non-functional requirements: Security, Reliability, Performance, Cost Optimization and Operationa…

Lucas looks at what can be learned from successes and failures in the world of distributions, and maybe transferred to language package managers. Topics discussed include Quality Assurance tools, collaboration with derivative distributions and upstream projets, and using distributions tools to deploy your own software.

Continuous deployment could be a nightmare, even when using the right tools. There's many of them available: CI servers, Provisionner, Containers… If you have to deal with feature-flipping too, it starts to be Hell. What if we try to get back to the basics and use only one tool, Git, for every of our tasks: from versioning to deploying? Matthias shows which …

Put CPU next to hard drives or SSDs for a new approach to large scale out infrastructures, improving overall efficiency while reducing failure domains. Enrico describes how to overcome the limitations imposed by fat serves and how to scale storage and compute for next generation workloads. The concept of offloading some CPU tasks to the storage infrastructur…

Serverless is being heralded as the next big thing in computing - but as you'll discover in this lightning talk, there's nothing new under the sun. The architectural underpinnings, technological constraints, and incumbent providers are all familiar. Practically speaking, Serverless still requires you to understand your software, your workload, and your use-c…

Willy covers multiple aspects of observability using the HAProxy load balancer. He also tries to suggest the smallest set of very relevant metrics to watch in order to detect when something starts to go wrong, and immediately spot what, where and help figure why.

TimescaleDB offers three seemingly heretical insights about data: (1) all data is time-series; (2) what you've heard about it is wrong; and (3) SQL should and does remain the lingua franca for databases. These insights enable new possibilities for architecting database systems, and Mike briefly discusses how TimescaleDB, an open-source time-series database p…

Our infrastructures generate a lot of metadata. However, true insight comes from the derivative data -- the data that we derive from the metadata. Jeremy gives you tips and strategies for collecting good metadata and how to use that to generate unique and actionable insights about your infrastructure.

Technology has gone through many iterations - Mainframes, to Client-Service Models, and now Isomorphic Applications. The world of Data Management has evolved too, from Databases, to Warehouses, Sharehouses, and Data Lakes, consequently Data Catalogues. Each of these iterations have given us learnings, and paved the way for better solutions! The constantly ch…

The Call of Duty game franchise generates a massive amount of data telemetry. Activision has been developing its data pipeline for a while, and they learned how to scale it not only in terms of traffic, but also in terms of supporting more games and more use-cases. Yaroslav dives into their unified message envelope and why it's so important to have a proper …

Istio, we have a problem! We've just deployed a shiny new set of micro-services and it behaves in a strange manner. Hard to say why with so many moving parts... Let's leverage the newly installed service mesh to understand what we've deployed. Find the root problem. Fix it. And then do a proper, non trivial, blue-green deployment of the updated version. That…