GitOps is a paradigm or a set of practices that empowers developers to perform tasks that typically (only) fall under the purview of operations. It’s a way to do Kubernetes cluster management and application delivery by using Git as a single source of truth for declarative infrastructure and applications. Being Git at the center of delivery pipelines, engineers use familiar tools to make pull requests to accelerate and simplify both application deployments and operations tasks to Kubernetes.

GitOps software agents (e.g. ArgoCD, Flux and Jenkins X) can alert on any divergence between Git with what's running in a cluster, and if there's a difference, Kubernetes reconcilers automatically update or rollback the cluster depending on the case.

This talk will include a demo of ArgoCD/Flux/Jenkins X on how to configure and use it to accelerate and simplify application deployments.


Outline/Structure of the Talk

Talk will outline the basics of DevOps, guide through some practical examples and demo how it can be achieved.

Learning Outcome

Understand what GitOps is and how it can help.

Target Audience

Software Engineers, SREs

schedule Submitted 3 months ago

Public Feedback

  • Yasunobu Kawaguchi
    By Yasunobu Kawaguchi  ~  3 months ago

    Thank you for submitting proposals.
    We have reorganized the topic categories to make them easier to understand.
    This is also reflected in the topics of this proposal.

  • クリス Chris Lucian

    クリス Chris Lucian - Mob DevOps & Mob Programming

    45 Mins

    Currently in our environment of 27 developers we have no manual tests for the software to deploy to production. We practice continuous integration, and continuous delivery. All infrastructure is written as code. Bugs and defects are top priority with no bugs most of the time. Team members treat each other with psychological safety. We have a 2-hour value stream for our cloud based IOT product. This means a customer can receive changes to the software within 2 hours of the start of the feature.


    How did we get to this point? By working collaboratively and practicing Mob DevOps and Mob Programming! In this session we will review the status of the team and how you can iterate toward a similar environment.

  • Hitesh Mandloi

    Hitesh Mandloi - Agile process improvements with Quality Control circle (QCC) and DevOps Automations

    20 Mins

    As team grows together with services, multiple challenges appear during team's daily operation, ranging from the number of critical and blocker issue reported by QA to adaptation of (and to) agile processes.

    Here we will share about Quality control, QA-automation, CI-CD activities done in our team.


    Problems :

    As a part of global team, Have you faced issues like below?

    - Not able to complete stories in sprint?

    - Members are not taking part actively in story planning, grooming?

    - Complexity to manage tasks / time in global team?

    - Regression testing takes man hour and time?

    - Afraid to do operations on production?

    - So many manual work, having mistakes frequently?

    - Facing infrastructure / server issues?


    Strategies :

    We have implemented some activities inside Rakuten group to overcome these issues.

    Here, You may learn

    - Develop, test, deploy planned stories to staging within sprint scope.

    - Planning with separate ticket for each single tasks

    - Grooming with poker game

    - Changing strategy of retrospective and action plan

    - FDD and QA automation with non-stop releases of Micro-services

    - Automations for repetitive / manual work

    - CI pipelines with Jenkins Shared Libraries, containers & deployment on K8s etc.


    Please join this session If you are interested in automations and improving agile processes.

  • Kazuki Suda

    Kazuki Suda / Masao Sakata - PFN の ML/DL 基盤を支える Kubernetes における自動化

    45 Mins

    Preferred Networks(PFN)は深層学習などの最先端の技術を最短路で実用化することで、これまで解決が困難であった現実世界の課題解決を目指しています。コンピュータビジョン、自然言語処理、音声認識、ロボティクス、コンパイラ、分散処理、専用ハードウェア、バイオインフォマティクス、ケモインフォマティクスといった幅広い分野で研究開発を行っており、それを支えているのが Kubernetes を用いて構築しているオンプレミス/ベアメタルの GPU クラスタです。

    本セッションでは、PFN が Kubernetes を用いてクラスタを運用するなかでどのような障害が起きるのかを紹介し、また障害対応をどのように自動化しているのかを具体的に使用/開発したソフトウェアを含めてご紹介します。また Kubernetes クラスタの管理、アップグレードの自動化にも取り組んでおり、それを実現する Cluster API についてもご紹介します。

  • Masato Wakahara

    Masato Wakahara / Ryusuke Tanaka - 決済戦国時代において、早く安全なシステムを提供するために with Azure

    45 Mins



  • 45 Mins

    What is a service mesh and when is it useful?

    A service mesh is a dedicated infrastructure layer for making service-to-service communication safe, fast, and reliable. If you’re building a cloud-native application, you need a service mesh. It can help connect, secure, control, and observe services. At a high level, a service mesh helps reduce the complexity of these deployments and ease the strain on your development teams.

    How does Linkerd/Istio provide observability by collecting metrics and telemetry about services? How does Linkerd/Istio use mTLS to create secure lines of communication between services? How does Linkerd/Istio improve reliability by using retries, timeouts, load baling, and traffic shifting?