Sao Paulo, San Pablo, Brasil
884 seguidores Más de 500 os

Unirse para ver el perfil

Acerca de

As a Senior Site Reliability Engineer, I aim to build sustainable internal development…

Actividad

Unirse para ver toda la actividad

Experiencia y educación

  • Nubank

Mira la experiencia completa de Gabriel

Mira su cargo, antigüedad y más

o

Al hacer clic en «Continuar» para unirte o iniciar sesión, aceptas las Condiciones de uso, la Política de privacidad y la Política de cookies de LinkedIn.

Publicaciones

  • SRE & Cloud & Containers (Presentation)

    Meetup "Engenho do Software #2"

    Introduction to DevOps, SRE, Cloud Architectures and containers using AWS, Docker, Kubernetes and commenting about trending technologies. In short: a presentation about hype.

    https://www.meetup.com/Engenho-de-Software/events/261437906

    Ver publicación

  • Sailing Delivery (Presentation)

    TDC Porto Alegre

    Solving continuous delivery inside Large Scale Multicloud/Kubernetes environments using "Spinnaker", an open source tool developed by Netflix.

    http://www.thedevelopersconference.com.br/tdc/2018/portoalegre/trilha-devops-tools

    Ver publicación

Cursos

  • Advanced Architecting on AWS

    -

Proyectos

  • Kubernertes implementation from scratch

    - actualidad

    We worked together with a graphics company through this project.

    Some of their developers were breaking a monolithic PHP application into microservices.
    We were tasked to find the best solution to solve this problem. Including implementation and of:
    - CI/CD Pipelines
    - Infrastructure
    - Authentication/Authorization of developers
    - Monitoring

    We were biased towards Kubernetes because they plan to deploy every new microservice inside a centralized plataform…

    We worked together with a graphics company through this project.

    Some of their developers were breaking a monolithic PHP application into microservices.
    We were tasked to find the best solution to solve this problem. Including implementation and of:
    - CI/CD Pipelines
    - Infrastructure
    - Authentication/Authorization of developers
    - Monitoring

    We were biased towards Kubernetes because they plan to deploy every new microservice inside a centralized plataform. Kubernetes seemed like the best approach because:
    - Community
    - Open Source
    - Easy to manage
    - Scalable
    - Multicloud (no Vendor lock-in)
    - Most popular and "main" orchestrator at the time

    CI/CD PIPELINES
    I managed to think about a solution to empower developers to create pipelines as soon and easy as possible. There is the design below:

    - A "Helm Chart Repository", which stored default charts for languages "nodejs", "php", etc.
    - An "Application Model Language" written in YAML inside the repository's root, storing variables like "Which laguage is used?", "What is the application name?", "Should I use a local chart?", "Where is the Dockerfile?"
    - A Jenkinsfile Shared Library to centralize and version all the functions and methods to not repeat code
    - A Jenkinsfile DSL method, which was separated by language and spawned a pipeline with all the necessary methods. You can build something similar following this link (https://jenkins.io/blog/2017/10/02/pipeline-templates-with-shared-libraries/). An example below:

    nodejs {
    configYaml = "myApplicationDeclaration.yaml"
    valuesYaml = "myHelmChartValues.yaml"
    }

    This was enough to deliver their NodeJS applications to any Kubernetes environment without needing to call the Ops team to create pipelines. We created once, used for many projects and decoupled from the application repositories.

    We also implemented:
    - Heptio Authenticator to manage RBAC
    - Prometheus Operator for monitoring
    - Consul+Vault to manage app secrets

    Otros creadores

  • Panoptes

    - actualidad

    Open source Python project, focused on analyzing multicloud security groups

    Ver proyecto

  • Reducing CI/CD build time

    -

    Four applications were built on Docker Cloud and deployed inside a Rancher Cattle cluster. All this process took around 20 minutes to finish.

    I was tasked to reduce the amount of build time from 30 minutes to acceptable 5 minutes.
    I accomplished it by creating a PoC with a Jenkins pipeline, which needed native Blue Ocean at the time, so no plugins like "Extensible Choice Parameter" were allowed.

    We found that ECS could be a better option to manage versioning and…

    Four applications were built on Docker Cloud and deployed inside a Rancher Cattle cluster. All this process took around 20 minutes to finish.

    I was tasked to reduce the amount of build time from 30 minutes to acceptable 5 minutes.
    I accomplished it by creating a PoC with a Jenkins pipeline, which needed native Blue Ocean at the time, so no plugins like "Extensible Choice Parameter" were allowed.

    We found that ECS could be a better option to manage versioning and deployments as well, so we planned to migrate those applications to AWS services.

    We used ECS Fargate for backend applications (NodeJS and PHP) and S3(React and Angular) for frontend applications. As all of the fronted could be generated statically, we chose S3 instead of containers which ended cheaper and more scalable for the company.

    I wrote the first pipeline skeleton, which built the image, stored on ECR and deployed on ECS Fargate. After some discussion and commits, I added BlueOcean Tests by outputting test results to JUnit XML and publishing them through the pipeline.

    After the first pipeline skeleton finished, I achieved the required 5 minutes of build time.

    For the frontend, I got the same skeleton but modified the Build and Deployment commands to copy the packed files to S3.

    After a few months, Diego Gomes got the task to migrate the 3rd and 4th application.
    He managed to tune the pipeline and replicate the same build performance for these applications, which later he migrated out of Rancher to ECS Fargate and finally moved the entire stack out of Rancher.

    Achievements:
    - Reduced build time from 30 minutes to 5 minutes
    - Reduced Frontend infrastructure costs by moving out of containers to S3 static files

    Otros creadores
    Ver proyecto

Idiomas

  • Português

    Competencia bilingüe o nativa

  • English

    Competencia profesional completa

  • Español

    Competencia básica

Más actividad de Gabriel

Ver el perfil completo de Gabriel

  • Descubrir a quién conocéis en común
  • Conseguir una presentación
  • ar con Gabriel directamente
Unirse para ver el perfil completo

Perfiles similares

Otras personas con el nombre de Gabriel Tiossi

Hay 3 personas más con el nombre de Gabriel Tiossi en LinkedIn

Ver a otras personas con el nombre de Gabriel Tiossi

Añade nuevas aptitudes con estos cursos

Ver todos los cursos