Compute Overview
This topic provides an overview of using Megaport ONE compute resources to run cloud-native applications and containerized workloads in Kubernetes clusters.
Using Megaport ONE compute as a managed service removes the complexity of Kubernetes cluster operation and maintenance, freeing you to focus on running your core business.
Benefits
-
Reduce the time to solution. Easy initial setup configures a fully working, secure cluster in a few minutes.
-
Deploy production-ready, cloud-native applications using Helm chartsHelm is an essential application package manager that describes the structure of an application in a packaging format called charts. A Helm chart contains a bundle of the YAML configuration files needed to run an application, including dependencies and libraries..
-
Standardize your deployment processes and share applications. This portable application packaging approach includes high availability, scalability, geo load balancing, fault isolation, and security.
-
Focus on modernizing applications. Megaport ONE Kubernetes cluster hosting simplifies the ongoing administration and maintenance of clusters, freeing you to focus on building applications. You can modernize existing applications, build in resiliency, and deliver them continuously.
-
Launch GPU-destined workloads using GPU resources. Bring the workload down when compute is finished (or analysis is complete), paying only for what you use. Running high-performance computing for a specific workload gives you the flexibility to change resources easily and as often as your workload requires.
Compute workflow
-
Create a cluster to receive the application. For details, see Creating a Cluster.
-
Collect the application package.
-
Search the application repository for applications. The application repository is a location that stores and shares applications prepackaged in Helm charts.
-
Use a Helm chart template to package an application. You can package your own containerized application in Helm charts. Megaport ONE provides a templated approach that makes it easy to create and initialize a chart. Partners can use this feature to bundle applications and host them for their managed users.
-
-
Deploy the application to a cluster. Deploying an application containerizes it, pushes its Helm chart to the repository, and automatically executes its deployment script, making it available to others. You can securely run a variety of applications on a new or existing Kubernetes cluster, and you can add applications to multiple clusters. The application is exposed to your users when they view the cluster or application repository. For details, see Viewing and Managing Applications.
-
Customize the application as needed. Read the chart’s
values.yamlfile, and then replace the default values with your own to further customize the deployment. Use system variables to pass specific values to an application. For details, see Viewing and Managing Applications.