+++
title = "Kubeflow on Google Cloud Platform"
+++

<!-- Hero section -->
{{<blocks/cover2 title="Kubeflow on Google Cloud" image_anchor="center" color="dark">}}
<div class="mx-auto">
  <p class="lead mt-5">The Machine Learning Toolkit running on Google Kubernetes Engine (GKE)</p>
  <a
    class="btn btn-lg btn-primary mr-3 mb-4"
    href="https://googlecloudplatform.github.io/kubeflow-gke-docs/docs/deploy"
  >
    Get Started <i class="fas fa-arrow-alt-circle-right ml-2"></i>
  </a>
</div>
{{</blocks/cover2>}}

<div id="overview">
  <div class="contain">
    <h4 class="section-head">What is Kubeflow?</h4>
    <p>
      Kubeflow on Google Cloud is an open-source toolkit for building
      machine learning (ML) systems. Seamlessly integrated with GCP services
      Kubeflow allows you to build secure, scalable, and reliable ML workflows of
      any complexity, while reducing operational costs and development time.
    </p>
  </div>
</div>

<section id="pageContent">
  <div class="container">
    <div class="card-deck">
      <div class="card border-primary-dark">
        <img
          src="./docs/images/logos/gke.png"
		  class="card-img-top"
		  draggable="false"
          style="padding: 2rem;"
          alt="Google Kubernetes Engine"
        />
        <div class="card-body bg-primary-dark">
          <h5 class="card-title text-white section-head">Google Kubernetes Engine</h5>
          <p class="card-text text-white">
            Kubeflow on Google Cloud uses the most automated and scalable managed Kubernetes platform, <a target="_blank" rel="noopener" href="https://cloud.google.com/kubernetes-engine">Google Kubernetes Engine</a>. By choosing Google Kubernetes Engine,
            one can benefit from advanced cluster managements features including automated node pool scaling,
            automatic software upgrades, node auto-repair, logging and monitoring of the cloud resources, and many more.
          </p>
        </div>
      </div>
      <div class="card border-primary-dark">
        <img
          src="./docs/images/logos/kfp.png"
		  class="card-img-top"
		  draggable="false"
          style="padding: 2rem;"
          alt="Kubeflow Pipelines"
        />
        <div class="card-body bg-primary-dark">
          <h5 class="card-title text-white section-head">ML Pipelines</h5>
          <p class="card-text text-white">
			      Kubeflow Pipelines is a comprehensive solution for deploying
            and managing end-to-end ML workflows. Use Kubeflow Pipelines for rapid and reliable experimentation.
            You can schedule and compare runs, and examine detailed reports on each run.
          </p>
        </div>
      </div>
      <div class="card border-primary-dark">
        <img
          src="./docs/images/logos/gcp.png"
		  class="card-img-top"
		  draggable="false"
          style="padding: 2rem;"
          alt="GCP's managed services"
        />
        <div class="card-body bg-primary-dark">
          <h5 class="card-title text-white section-head">
            Managed services
          </h5>
          <p class="card-text text-white">
            Tightly integrated with managed services, Kubeflow on Google Cloud allows to work faster, build smarter, and grow with a peace of mind.
            Managed services include Cloud Storage, CloudSQL, Identity and Access Management, Identity-Aware Proxy, Anthos Service Mesh, and many others.
          </p>
        </div>
      </div>
    </div>
    <br />
    <!-- For future editors: If you are adding cards not in multiples of 3, wrap them in a 'card-columns' div 
						instead of card-deck. It will look more pleasing to the eye. -->
  </div>
</section>

<div id="community">
  <div class="contain">
    <h4 class="section-head">Source code</h4>
    <p>
      Kubeflow on Google Cloud is open-source. 
      <a class="btn btn-lg btn-primary mr-3 mb-4" href="https://github.com/googlecloudplatform/kubeflow-distribution">
        Explore and contribute!
        <i class="fas fa-arrow-alt-circle-right ml-2"></i>
      </a>
    </p>
    <p>
      Interested in getting more involved? <a href="https://www.kubeflow.org/docs/about/community/">Join Kubeflow Community!</a>
    </p>
  </div>
</div>