Help Center> GaussDB(DWS)> Billing> Cost Management
Updated on 2023-09-20 GMT+08:00

Cost Management

As you migrate more of your services to the cloud, managing cloud costs becomes more important. Cost management can reduce your financial burden. The following describes how to manage costs from four dimensions: cost composition, allocation, analysis, and optimization, to help maximize return on your investment.

Cost Composition

The cost of using GaussDB(DWS) includes the following two aspects:

  • Resource cost: the cost of each type of resources and resource packages, which depends on the GaussDB(DWS) billing items. For details, see Item.
  • O&M cost: the labor cost generated when GaussDB(DWS) is used.
Figure 1 GaussDB(DWS) costs

Cost Allocation

A good cost accountability system is the basis of cost management. It ensures that departments, business teams, and owners are accountable for their respective cloud costs. Allocate costs to different teams or projects so that organizations have a clear picture of their respective costs.

Huawei Cloud Cost Center supports cost collection and reallocation with multiple tools for you to choose from.

  • Allocate costs by linked account.

    The enterprise master account can categorize the costs of its member accounts by linked account to manage the accounting of those member accounts. For details, see Viewing Costs by Linked Account.

  • Allocate costs by enterprise project.

    Before allocating costs, enable Enterprise Project Management Service (EPS) and plan your enterprise projects based on your organizational structure or businesses. Select an enterprise project for a newly purchased cloud resource so that the costs of that resource will be allocated to the selected enterprise project. For details, see Viewing Costs by Enterprise Project.

  • Allocate costs by cost tag.

    Huawei Cloud assigns tags to your cloud resources so they can be sorted in different ways, for example, by purpose, owner, or environment. The following is the process of managing costs by predefined tags.

Figure 2 Adding tags to GaussDB(DWS)

For details, see Viewing Costs by Cost Tag.

  • Allocate costs by cost category.

    You can use Cost Categories provided by Cost Center to split shared costs. Shared costs include the costs for the resources (compute network, storage, or resource packages) shared across departments or the costs that cannot be directly split by cost tag or enterprise project configured for the resources. These costs are not directly attributable to a singular owner, and hence cannot be categorized into a singular cost category. In this case, you can define cost splitting rules to fairly allocate these costs among teams or business units. For details, see Viewing Cost By Cost Category.

Cost Analysis

To accurately control and optimize your costs, you need a clear understanding of what parts of your enterprise incurred different costs. Cost Center visualizes your original costs or amortized costs using various dimensions and display filters for cost analysis so that you can analyze the trends and drivers of your service usage and costs from a variety of perspectives or within different defined scopes.

You can also use Cost Anomaly Detection provided by Cost Center to detect unexpected expenses in a timely manner. In this way, costs can be monitored, analyzed, and traced.

For details, see EPerforming Cost Analysis to Explore Costs and Usage and Enabling Cost Anomaly Detection to Identify Anomalies.

Cost Optimization

  • Cost control

    You can create different types of budgets on the Budgets page of Cost Center to track your costs against the budgeted amount you specified and send alerts to the recipients you configured if the thresholds you defined are reached. You can also create budget reports and Huawei Cloud will periodically generate and send to the recipients you configured on a schedule you set.

    Suppose that you want to create a monthly budgeted amount of 2000 USD for a pay-per-use GaussDB(DWS) instance, and to receive an alert if the forecasted amount exceeds 80% of the budgeted amount. Then, you can create a budget as follows:

    Figure 3 Basic budget information

Figure 4 Budget scope

Figure 5 Budget alerts

For details, see Enabling Forecasting and Creating Budgets to Track Cost and Usage.

  • Resource optimization

    Cost Center analyzes your historical CPU and other resource usage to identify idle ECS resources and generate resource rightsizing recommendations. These rightsizing opportunities are places where you can reduce costs. You can also identify resources with high costs based on the analyses on the Cost Analysis page and use Cloud Eye to monitor resource usage. By doing this, you can determine the causes of high costs and take optimization measures accordingly. You can:

    • Monitor resource usage and evaluate whether the current configuration is too high. For example, you can monitor the usage of CPUs, memory, cloud disks, and bandwidth.
  • Billing mode selection

    Different types of services have different requirements on resource usage periods and therefore require different billing modes to achieve the optimal outcome.

    • For mature services that are stable for a long time, use the yearly/monthly billing mode.
    • For short-term, unpredictable services that experience traffic bursts and cannot be interrupted, use pay-per-use billing.
  • O&M automation

    Huawei Cloud also provides various O&M products to help you improve O&M efficiency and reduce O&M labor costs. For example:

    Cluster scale-out: Add nodes to an existing cluster. The storage size of the added nodes is the same as that of the existing nodes in the cluster. It applies to scenarios where both CPU and storage are insufficient.

    Cluster scale-in: Delete nodes from an existing cluster. It applies to scenarios where both computing and storage resources are redundant.

    Resize: Upgrade or downgrade the specifications of an existing cluster or migrate the cluster to a new cluster with new specifications. Specification upgrade or downgrade is applicable to scenarios where CPU bottlenecks or redundancy exists. Migration is applicable to re-planning of the entire cluster.

    Disk scale-out: Expand the capacity of an existing cluster disk. It applies to scenarios where the cluster storage space is insufficient.