liveBook · Manning

Managed service

As explained in Platform Engineering on Kubernetes

A managed service is a provider-operated service that abstracts the administration of its software, infrastructure, and operational components.

A managed service is a provider-operated service that abstracts administration of software, infrastructure, and operational components. Higher-level services absorb more underlying layers and operational costs, increasing convenience and price. section 1.1.1

For a highly available PostgreSQL instance, the provider manages the database software, operating system, hardware, and monitoring tools. Teams trade direct control and operating effort for a service charge. section 1.1.1

A managed Kubernetes offering shifts cluster operations while introducing practical constraints:

  • Provision real clusters through a self-service cloud account, scale resources for a task, and delete clusters when finished. section 2.1.1
  • Budget caps and access controls become necessary because consumption is billed by the provider. section 2.1.1
  • Remote environments can slow local workflows, but they also make development and deployed applications behave more like production. section 2.1.1
  • Evaluate provider dependence and organizational policies before placing workloads or data in the cloud. section 1.1.1 section 2.1.1
Figure 1.1 Cloud provider’s services categories

Use the service-category diagram to compare low-level infrastructure services with higher-level application services. section 1.1.1

Figure 1.2 Provisioning a PostgreSQL database instance in the cloud

Use the PostgreSQL provisioning figure to inspect which database, operating-system, hardware, and monitoring responsibilities the provider absorbs. section 1.1.1

How Platform Engineering on Kubernetes covers it

How deep it goes
Explains bundled responsibilities, costs, benefits, and constraints, but stops short of provider-specific configuration procedures.
Examples
Uses a highly available PostgreSQL instance and managed Kubernetes offerings such as GKE, AKS, and EKS.
What it assumes
Assumes a cloud-provider account and attention to budgets, access, workload location, and organizational policies.
What you can do after
Operate managed environments by provisioning clusters, allocating resources, applying access and budget controls, and deleting unused resources.
How it connects
Connects managed services with cloud providers, PostgreSQL, Kubernetes, monitoring, remote environments, and vendor lock-in.

Also explained in

Compare how 3 books explain managed service →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Platform Engineering on Kubernetes. The section links point to the parts of the book it draws on.