Este conteúdo não está disponível no idioma selecionado.
Chapter 1. Why use cost management
Cost management is a free offering as part of your subscription to the Red Hat Lightspeed portfolio of services. With cost management, you can monitor and analyze your costs to improve the management of your business.
Cost management helps you simplify the management of your resources and costs across container platforms like OpenShift Container Platform, as well as public clouds like Amazon Web Services (AWS), Google Cloud, and Microsoft Azure.
1.1. What can you accomplish with cost management? Copiar o linkLink copiado para a área de transferência!
With the expanding scale and performance of containerized business applications, you need aggregated and meaningful data so that you can quickly analyze your cluster spending and align with business priorities.
To overcome business challenges, cost management gives your organization visibility into your costs down to the project level for on-premise and public cloud environments. This visibility gives IT and financial stakeholders a unique snapshot into the costs associated with applications.
With cost management, you can achieve some of the following goals:
- Visualize, understand, and analyze how you use your resources and costs across hybrid cloud infrastructure
- Track cost trends
- Map charges to projects and organizations
- Use cost models to normalize data and add markups
- Forecast your future consumption and compare it with your budgets
- Optimize your resources and usage
- Identify patterns of usage that you might want to investigate
- Integrate with third party tools that can use your cost and resourcing data
These preceding goals can ultimately help your organization optimize costs, increase efficiency, and save money.
1.2. Key OpenShift concepts Copiar o linkLink copiado para a área de transferência!
It is important to understand the following key OpenShift concepts to work with cost management:
- Cluster
- a group of servers that are managed together and participate in workload management.
- Node
a worker machine that is either virtual or physical, depending on the cluster.
- Master node: The master node hosts the control plane and manages the cluster, including scheduling and scaling applications and maintaining the state of the cluster.
- Worker node: Worker nodes are responsible for running the containers and executing the workloads.
- Pod
- a collection of one or more containers. It is the smallest unit possible.
- Persistent volume claim (PVC)
- Persistent volume (PV) framework enables cluster administrators to provision persistent storage for a cluster. Developers can use persistent volume claims (PVCs) to request PV resources.
1.3. How does cost management work? Copiar o linkLink copiado para a área de transferência!
Cost management distributes the sum of the costs from your cloud provider (if any) and the costs in any price lists. Price lists can be defined in Red Hat OpenShift cost models and there are more than 40 dimensions that you can use to define a unit cost. Your environment will have one of the following configurations:
- An on-premise environment with only custom-defined price lists
- A cloud environment with no custom-defined price lists
- A cloud environment that also has one or more custom-defined price lists
The following steps outline how cost management distributes the costs for a cloud environment, either with or without custom-defined price lists:
From your cloud bill, cost management takes the cost of all of your nodes and determines what nodes belong to what cluster and which nodes are worker or master nodes.
NoteIf you are a Red Hat OpenShift Service on AWS (ROSA) or an Azure Red Hat Openshift customer (ARO), the cost of your ROSA or ARO subscription is included in the cloud bill and cost management automatically distributes it to your workloads. However, due to limitations in the data that AWS accepts from the Red Hat Subscriptions service, for Red Hat OpenShift on AWS Hosted Control Plane (ROSA HCP) subscriptions, you must create a cost model that includes some rates, as explained in the ROSA HCP clusters may not show their full cost in Red Hat Insights cost management knowledge base article. Red Hat is working to resolve this issue.
- Cost management then determines what pods were running on what node of what cluster at each moment in time, and calculates how much central processing units (CPU), memory, GPU, storage, network traffic, and so on that each pod used.
- Cost management multiplies the cost from the cloud bill by the established usage metrics to calculate the amount of money that each pod is incurring at each moment in time.
Cost management applies a cost model to distribute the cost of the platform or the cost of unallocated capacity.
If you did not create a Red Hat OpenShift Container Platform cost model, we use the default cost model which distributes the cost from the cloud bill based on CPU effective use.
Cost management does not use public prices. Instead, it reads your cloud bill to process the savings plans, reserved instances, discounts, or other costs that you have. Cost management also tracks which pods run on which nodes at each moment in time. If you have different instance types, or same instance types but with different prices, cost management can still attribute the correct cost to each pod.