ACK Cost Insights Overview: FinOps Visibility for Kubernetes Clusters - Container Service for Kubernetes

The cost insights feature of Container Service for Kubernetes (ACK) helps IT administrators of enterprises analyze the usage and cost allocation of resources in ACK clusters and provides suggestions on cost savings to improve the overall resource utilization. After you enable the cost insights feature, you can gain insights into the cost and resource usage of a specified cluster, department, or application within a specified financial governance cycle. This meets the requirements for cost estimation, allocation, and accounting in various scenarios.

Prerequisites

You have connected Alibaba Cloud ACK to Cloud Monitor 2.0.

Purpose of cost insights

Large numbers of enterprises use cloud-native technologies for IT transformation in which cost optimization is an important objective. You can use the sharing, isolation, and scaling capabilities of cloud-native technologies to optimize the costs of resources. Compared with the traditional capacity planning method that is used to manage IT spending, cloud-native technologies pose greater challenges. The following section describes these challenges:

How to accurately calculate the cost of an ACK cluster?
How to accurately estimate the cost of a pod?
How to accurately estimate the cost of an online application or a CronJob?
How to use namespaces to allocate the cost of ACK cluster management by department?
How to identify and optimize cost waste in a cluster in a visualized manner?

Figure 1. Resource utilizations of different types of clusters

To address the preceding challenges, ACK provides the cost insights feature. The cost insights feature is an important part of Finance+DevOps (FinOps) and plays an important role throughout the entire cost governance procedure. For example, you can use this feature to monitor the daily cost trend, analyze cost issues, and evaluate cost optimization measures.

Dimensions of the Cost Insights dashboard

Dimension	Description
CS Cost Overview	Provides insights into the overall resources and costs of a cluster, reflecting the health of your cluster costs. For any cluster, first check the overall cost trend to see if it meets your expectations. If the trend is abnormal, use the dashboard data to find the root cause.
CS Cost Namespace	If your company uses namespaces to separate departments or services, you can filter by namespace to view resource and cost information for a specific one.
CS Cost Node Pool	Provides cost insights from a cluster resource perspective. Computing resources, mainly ECS instances, are often the largest part of cluster costs. O&M engineers manage these resources directly. The node pool cost insights provide analysis of node pool resource usage and help you choose a billing strategy.
CS Cost Application	Focuses on scenario-based cost optimization. Use a label selector (a label with a wildcard character) to filter for specific applications and collect cost and resource statistics. With a label selector, you can monitor the cost and resource usage of a single application or multiple related applications. For example, in a big data workflow, you can add the same label to all applications in the pipeline. This lets you analyze the cost of the entire workflow.

Access the feature

Log on to the Cloud Monitor 2.0 console, and select a workspace. In the left navigation pane, choose Application Center > CloudLens > CloudLens App For Container.
In the navigation bar of CloudLens for Container, choose Insight > Cost Insights.
On the dashboard page, click the tabs to view the cost visualization dashboards.
- CS Cost Overview
- CS Cost Namespace
- CS Cost Node Pool
- CS Cost Application

CS Cost Overview: Cluster cost analysis

Filter dimensions

Parameter	Description
Cluster billable cost	The cluster cost metric options. You can view statistics for the actual billable cost after discounts or the list price billable cost. Actual Billable Cost (after Discounts): The default option for the dashboard. The cost statistics show the payable amount for all cloud resources in the cluster. List Price Billable Cost: The cost statistics show the list price for all cloud resources in the cluster. For more information about payable amounts and list prices, see Billing details (offline and redirected). Note Cost statistics for some applications in the cluster, such as those at the namespace and pod levels, are calculated based only on the list price billable cost.
Cost allocation model	The cost allocation model options. You can choose a single-resource model or a weighted hybrid-resource model. The models are as follows: CPU Model: The default option for the dashboard. Pod costs are estimated based on CPU resource requests. Memory Model: Pod costs are estimated based on memory resource requests. CPU-memory Hybrid Model (recommended Weights): Pod costs are estimated based on weighted CPU and memory metrics. This model uses system-recommended weights. CPU-memory Hybrid Model (custom Weights): Pod costs are estimated based on weighted CPU and memory metrics. To use this allocation model, first set the allocation model to CPU-memory Hybrid Model (custom Weights), and then edit the CPU Weight Settings. For more information about choosing a cost allocation model, see Cost estimation strategies.
Time range	Select a time range for the dashboard. You can set this parameter to view cost and resource trends over different time ranges.

Cluster cost overview

Metric	Description
Current CPU weight Current memory weight	The data displayed here depends on the Cost Allocation Model that you select. By default, the dashboard uses the CPU Model, which estimates pod costs based on CPU resource requests.
Yesterday's cluster cost Day-over-day cost change Tomorrow's predicted cost Cumulative cost this week Cumulative cost this month Predicted total cost this month	These are cost statistics. Yesterday's cluster cost, day-over-day cost change, cumulative cost this week, and cumulative cost this month are the billable costs of the cloud resources in this cluster. Cumulative cost this week and cumulative cost this month are the billable costs for the calendar week and calendar month, respectively. Because bills are generated with a T+1 latency, the cumulative cost for the week is not displayed on Mondays, and the cumulative cost for the month is not displayed on the first day of the month. The day-over-day cost change is the change in yesterday's cluster cost compared to the day before: If the font color of the change percentage is green, the cost has decreased compared to the previous day. If the font color of the change percentage is red, the cost has increased compared to the previous day.
Cluster cost & capacity trend chart	This is a trend chart of cluster cost and capacity. The yellow curve represents the cost, and the blue curve represents the actual cluster capacity. Typically, the two curves are correlated. You can compare the correlation between the two curves. If you find that their trends are inconsistent, an anomaly may exist in the cluster's unit core cost. In this case, check for resources with excessively high costs.
Real-time cost estimation (by namespace) Cost allocation (by namespace)	Real-time cost estimation: The real-time estimated cost for each namespace in the cluster. The cost of a namespace is the sum of the estimated costs of all its pods. Cost allocation: The allocated cost for each namespace. This is the portion of the cluster's actual cost allocated to the namespace based on its estimated cost proportion. Note Namespace costs are calculated based on the list price billable cost. If a pod in a namespace does not have a Resource Request (CPU) configured, the pod is considered to have no declared cluster resource requirements and is not included in the namespace cost calculation. A cluster may have various node types with different specifications and billing methods. When you allocate bills by namespace, you cannot rely solely on the resource request value of each namespace. You must also consider the nodes on which the pods in the namespace are located. Cost Insights converts the real-time cost of each node. When calculating namespace-level costs, the formula is: Σ (Pod resource request / Node capacity) × Node unit price This method can accurately estimate namespace costs. Various delayed billing policies, such as user discounts, coupon deductions, and subscription plans, may cause a discrepancy between the namespace cost and the cluster's actual bill. However, you can achieve namespace-level bill allocation by multiplying the total cluster cost by the namespace's cost percentage.

Stability & efficiency analysis

Metric	Description
Stability/Performance risks Cluster CPU usage overview Cluster memory usage overview	Shows the number of pods for each Quality of Service (QoS) class in the cluster and the total resource usage.
Cluster pod resource utilization analysis	This provides basic information and resource utilization (Usage/Request) for all pods in the cluster. It also supports filtering and sorting. You can use this feature to view the workloads with the highest or lowest resource usage in the cluster.
Burstable Pod - Resource usage analysis	You can view the resource configuration of Burstable QoS class pods. Filtering and sorting are supported. You can use this feature to view the request and limit settings for resources such as CPU and memory for each Burstable pod. This helps you understand the cluster resources that the pod consumes and identify potential resource bottlenecks.
Best Effort Pod - Resource usage analysis	You can view the resource configuration of BestEffort QoS class pods. These pods generally have higher stability risks. You can filter and sort the list to check for any unexpected BestEffort pods and address them promptly to mitigate risks.

Note

For more information about the Stability & Efficiency Analysis feature, see Use Cost Insights to identify cluster resource risks.

Cluster cost analysis

Metric	Description
Cloud product cost analysis Actual cost trend (by cloud product)	A cluster can contain multiple cloud products. Different cloud products have different usage methods and billing models, which can lead to variations in costs. You can view the consumption of different cloud products through cost trends and combinations to make cost-related decisions.
Actual cost trend (by cluster)	Statistics on the daily total cost trend of the cluster.
Actual cost trend (by node pool)	Analysis of the node billable costs for each node pool or virtual node in the cluster.
Cost estimation analysis (by namespace) - List price billable cost	A cluster may have various node types with different specifications and billing methods. When you allocate bills by namespace, you cannot rely solely on the resource request value of each namespace. You must also consider the nodes on which the pods in the namespace are located. Cost Insights converts the real-time cost of each node. When calculating namespace-level costs, the formula is: Σ (Pod resource request / Node capacity) × Node unit price This method can accurately estimate namespace costs. Various delayed billing policies, such as user discounts, coupon deductions, and subscription plans, may cause a discrepancy between the namespace cost and the cluster's actual bill. However, you can achieve namespace-level bill allocation by multiplying the total cluster cost by the namespace's cost percentage.
Cluster computing resource request & utilization trend chart	Scenarios: Analyze whether resource waste exists in the cluster's usage level or capacity. When scenarios such as elastic scaling occur, the cluster usage level will fluctuate periodically. This chart helps you plan resource capacity. Trend chart description: The y-axis represents the total capacity of the cluster's computing resources (Capacity). This indicates the total amount of application resources that the cluster can support. The green column chart shows the total requested computing resources (Request) that have been allocated in the cluster for the current hour. The yellow column chart shows the computing resources actually used by processes in pod containers (Usage) for the current hour. This is the resource amount that applications actually use. Allocated but unused application resources = Green column chart - Yellow column chart Remaining allocatable cluster resources = y-axis - Green column chart Analysis flow: Unallocated resource waste: You can refer to the remaining allocatable cluster resources to use the unallocated and wasted resources in the cluster. You can adjust the resource request amount (Request) of pods in the cluster or decrease the quota. We recommend that you keep the remaining allocatable resources at about 20% of the total cluster resources. Allocated but unused resource waste: You can refer to the amount of allocated but unused resources in the cluster. You can use the ranking of wasteful applications and pods on the namespace dashboard to find applications with large resource allocations but low actual usage. Then, you can decrease the resource allocation amount (Request). Auto Scaling scenarios: In common scenarios where business loads fluctuate periodically, you can refer to the fluctuation level of the column chart to plan resource capacity and configure appropriate elastic scaling policies.
Cluster billing details - By cloud product Cluster billing details - By instance	A list of daily bills for the cluster, broken down by cloud product and cloud product instance.

CS Cost Namespace: Namespace cost analysis

The namespace dashboard lets you filter and display resource and cost information for each namespace. Namespaces often correspond to departments or teams in a company.

Filter dimensions

Parameter	Description
Namespace	Select the namespace of the cluster to analyze. The default is ALL, which means the entire cluster.
Cost allocation model	The cost allocation model options. You can choose a single-resource model or a weighted hybrid-resource model. The models are as follows: CPU Model: The default option for the dashboard. Pod costs are estimated based on CPU resource requests. Memory Model: Pod costs are estimated based on memory resource requests. CPU-memory Hybrid Model (recommended Weights): Pod costs are estimated based on weighted CPU and memory metrics. This model uses system-recommended weights. CPU-memory Hybrid Model (custom Weights): Pod costs are estimated based on weighted CPU and memory metrics. To use this allocation model, you must first set the allocation model to CPU-memory Hybrid Model (custom Weights), and then edit the CPU Weight Settings. For more information about choosing a cost allocation model, see Cost estimation strategies.
Actual/List price bill	The cluster cost metric options. You can view statistics for the actual billable cost after discounts or the list price billable cost. Actual Billable Cost (after Discounts): The default option for the dashboard. The cost statistics show the payable amount for all cloud resources in the cluster. List Price Billable Cost Statistics: The cost statistics show the list price for all cloud resources in the cluster. For more information about payable amounts and list prices, see Billing details (offline and redirected).
Time range	Select a time range for the dashboard. The default is the last 7 days. You can set this parameter to view cost and resource trends over different time ranges.

Cost overview

Metric	Description
CPU resource usage Memory resource usage GPU resource usage	You can use the CPU and memory metrics to determine resource waste in a namespace. The metrics are as follows: CPU resources CPU Resource Usage: The number of CPU cores currently consumed by the namespace. CPU Resource Requests: The number of CPU cores currently allocated to the namespace. Total CPU Resource Capacity: The total number of CPU cores in the cluster. CPU Resource Utilization: The ratio of the namespace's current CPU usage to its requested amount. Memory resources: Memory Resource Usage: The amount of memory currently consumed by the namespace. Memory Resource Requests: The amount of memory currently allocated to the namespace. Total Memory Resource Capacity: The total amount of memory in the cluster. Memory Resource Utilization: The ratio of the namespace's current memory usage to its requested amount. GPU resources: GPU Memory Usage: The amount of GPU memory currently consumed by the namespace. Allocated GPU Memory: The amount of GPU memory currently allocated to the namespace. Total GPU Memory Capacity: The total amount of GPU memory in the cluster. GPU Memory Utilization: The ratio of the namespace's current GPU memory usage to its allocated amount.
Real-time namespace cost estimation	The real-time estimated cost of the namespace, which is the sum of the estimated costs of all pods within it.
Namespace cost allocation	The allocated cost for the namespace. This is the portion of the cluster's actual cost allocated to the namespace based on its estimated cost proportion.

Cost details and trends

Metric	Description
Unit price per core-hour	Trend statistics for the price per CPU core per hour of the nodes where the pods are located.
CPU/Memory resource utilization trend	Trend statistics for CPU and memory resource utilization in the namespace.
CPU resource request/usage trend Memory resource request/usage trend GPU memory request/usage trend	The trend of resource allocation and actual consumption. The blue curve represents allocated resources, and the red curve represents actually consumed resources. When a pod is scheduled, the node pre-allocates a certain amount of resources for it. However, the resources actually consumed by the container process are usually different from the pre-allocated resources. This chart reflects the relationship between the two, which helps you optimize wasted resources.
Namespace estimated cost trend - List price billable cost Trend of namespace cost as a percentage of total cluster cost	Cost trend statistics for applications in the namespace. The total cost is the cost of the namespace within the specified time range, calculated based on the list price billable cost. The percentage of the namespace's cost relative to the total cluster cost shows the cost proportion of the selected namespace.
Pod resource request ranking	Statistical analysis of applications with high pod resource requests in the cluster, used for scenarios such as capacity planning.
Pod resource utilization ranking - Sorted by CPU utilization	You can view pod application replicas with low resource utilization to help you identify idle applications.
Pod idle resource ranking	The idle resource rate is the proportion of a pod's unused computing resources to its allocated resources. It reflects the pod's resource waste. By analyzing the pods with the most resource waste in a namespace, you can directly identify the main applications that cause resource waste in the namespace. This helps you analyze the reasons for the waste and design targeted resource optimization strategies.

CS Cost Node Pool: Node pool cost analysis

The node pool dashboard provides cost insights from a cluster resource perspective. It helps you analyze node pool resource usage and choose a billing strategy.
The node pool dashboard focuses on the resource dimension, which allows for resource cost planning and administration from the perspective of different node pools. For example, resources such as GPU node pools may belong to multiple departments, which makes it difficult to analyze costs by namespace. With the node pool dashboard, you can directly set policies and optimize costs from the resource dimension.

Filter dimensions

Parameter	Description
Actual/List price bill	The cluster cost metric options. You can view statistics for the actual billable cost after discounts or the list price billable cost. Actual Billable Cost (after Discounts): The default option for the dashboard. The cost statistics show the payable amount for all cloud resources in the cluster. List Price Billable Cost Statistics: The cost statistics show the list price for all cloud resources in the cluster. For more information about payable amounts and list prices, see Billing details (offline and redirected).
NodePoolID	Select the node pool of the cluster to analyze. The default is All, which means all node pools.
Time range	Select a time range for the dashboard. The default is the last 7 days. You can set this parameter to view cost and resource trends over different time ranges.

Cost overview

Metric	Description
Yesterday's node cost Day-over-day change in node pool node cost Node pool node cost as a percentage of total cluster cost Node unit price per core-hour - Entire cluster	If the font color of the change percentage relative to yesterday's cost is green, the cost has decreased compared to the previous day. If the font color of the change percentage relative to yesterday's cost is red, the cost has increased compared to the previous day.
Unit price per core-hour for the cluster	The trend of changes in the unit price per core-hour for nodes in the node pool over a period of time.
Node pool cost trend Trend of node pool node cost as a percentage of total cluster cost	The trend of cost changes for nodes in the node pool, and the trend of the proportion of these node costs to the total cluster cost.

Billing strategies and cost estimation

Metric	Description
Trend of the number of node instances for different billing strategies Cost trend of node instances for different billing strategies - List price billable cost Statistics of billing strategies for all cluster nodes - List price billable cost Statistics of the number of instances for different billing strategies for all cluster nodes	Analysis of the proportion and trend of node instances with different billing strategies over a period of time, along with the analysis of the cost proportion and trend of nodes with different billing strategies in the node pool. The following billing strategies are supported: PayAsYouGo: Pay-as-you-go. PayByPeriod: Subscription. Spot: Spot instance.
Predicted cost if all nodes are pay-as-you-go Predicted cost if all nodes are subscription Predicted cost if all nodes are Spot Predicted cost if all nodes are pay-as-you-go Predicted cost if all nodes are subscription Predicted cost if all nodes are Spot Daily cost estimation for node pool nodes with different billing strategies - List price billable cost	Cost savings prediction analysis for changing node billing strategies. This analyzes the potential savings or extra costs if all nodes in the current cluster's node pool are switched to different billing strategies, which helps you optimize your resource mix and billing strategies.
Hourly cost/billing strategy statistics for nodes - List price billable cost	Statistics on the cost and billing strategies for all nodes in the node pool.

CS Cost Application: Application cost analysis

The application dashboard uses label wildcard matching to filter applications that you follow. It provides cost and resource statistics for scenario-based cost optimization. Typical scenarios include big data services, AI services, and elastic services.
Using label wildcard matching, you can monitor the cost and resource usage of a single application or multiple related applications. For example, in a big data workflow scenario, you can add a consistent label to all applications in the pipeline to analyze the cost of the entire workflow.

Filter dimensions

Parameter	Description
Namespace	The cluster cost metric options. You can view statistics for the actual billable cost after discounts or the list price billable cost. Actual Billable Cost (after Discounts): The default option for the dashboard. The cost statistics show the payable amount for all cloud resources in the cluster. List Price Billable Cost Statistics: The cost statistics show the list price for all cloud resources in the cluster. For more information about payable amounts and list prices, see Billing details (offline and redirected).
Namespace	Select the namespace of the cluster to analyze. The default is All, which means all namespaces.
Workload type	The object type of the cluster resource.
Workload name	Based on the resource object type, select the workload name.
Filter by label pair (LabelSelector)	Enter the label of the application pod. You can use `=` or `:` to connect the label's key and value, for example, `app=ack-cost-exporter` or `app:ack-cost-exporter`. If the label key contains `/`, `.`, or `-`, you must convert these symbols to `_` and remove all `"` characters. For example: `"sparkoperator.k8s.io/submission-id":"db08a66a-c0b7-4d32-8013-02ac4f8eff4c"` needs to be converted to: `sparkoperator_k8s_io_submission_id:db08a66a-c0b7-4d32-8013-02ac4f8eff4c`
Cost allocation model	The cost allocation model options. You can choose a single-resource model or a weighted hybrid-resource model. The models are as follows: CPU Model: The default option for the dashboard. Pod costs are estimated based on CPU resource requests. Memory Model: Pod costs are estimated based on memory resource requests. CPU-memory Hybrid Model (recommended Weights): Pod costs are estimated based on weighted CPU and memory metrics. This model uses system-recommended weights. CPU-memory Hybrid Model (custom Weights): Pod costs are estimated based on weighted CPU and memory metrics. To use this allocation model, you must first set the allocation model to CPU-memory Hybrid Model (custom Weights), and then edit the CPU Weight Settings. For more information about choosing a cost allocation model, see Cost estimation strategies.
Time range	Select a time range for the dashboard. The default is the last 7 days. You can set this parameter to view cost and resource trends over different time ranges.

Cost overview

Metric	Description
Application cost	The cost of the currently selected application for the time range queried on the dashboard, calculated based on the list price billable cost.
Current number of application replicas	Statistical analysis of the peak and trough values of the pod replica count for the application.
Application resource utilization in the entire cluster/namespace	The resource proportion of the application in the cluster and namespace.
Hourly cost of the node where the application runs	The unit price per core-hour for each node where the application is running.
Application runtime / Total consumed core-hours	Statistics based on the application's runtime. The total number of core-hours consumed by the application.
Computing resource utilization	The utilization rate of CPU and memory (Usage / Request).
Pod-level application cost analysis	You can view the resource status and real-time estimated cost of each pod that is contained in the application.
Application estimated cost trend	The trend of changes in the application's hourly cost and unit price per core-hour over a period of time.
Application pod scale trend	Trend statistics for the number of pod replicas for the application.
CPU request & usage trend Memory request & usage trend GPU memory request & usage trend	Description of the CPU, memory, and GPU request and usage trend charts: The y-axis represents the total capacity of the cluster's computing resources (Capacity). This indicates the total amount of application resources that the cluster can support. The blue column chart shows the total requested computing resources (Request) that have been allocated in the cluster for the current hour. The yellow column chart shows the computing resources actually used by processes in pod containers (Usage) for the current hour. This is the resource amount that applications actually use. Allocated but unused application resources = Blue column chart - Yellow column chart Remaining allocatable cluster resources = y-axis - Blue column chart Waste analysis flow: Unallocated resource waste: You can refer to the remaining allocatable cluster resources to use the unallocated and wasted resources in the cluster. You can adjust the resource request amount (Request) of pods in the cluster or decrease the quota. We recommend that you keep the remaining allocatable resources at about 20% of the total cluster resources. Allocated but unused resource waste: You can refer to the amount of allocated but unused resources in the cluster. You can use the ranking of wasteful applications and pods on the namespace dashboard to find applications with large resource allocations but low actual usage. Then, you can decrease the resource allocation amount (Request). Auto Scaling scenarios: In common scenarios where business loads fluctuate periodically, you can refer to the fluctuation level of the column chart to plan resource capacity and configure appropriate elastic scaling policies.

Billing strategies and pod usage

Metric	Description
Predicted cost if all nodes are pay-as-you-go Predicted cost if all nodes are subscription Predicted cost if all nodes are Spot Predicted cost if all nodes are pay-as-you-go Predicted cost if all nodes are subscription Predicted cost if all nodes are Spot	Analyzes the potential savings or extra costs if all nodes in the current application are switched to different billing strategies, which helps you optimize your resource mix and billing strategies.
Billing strategy statistics for nodes where the application runs Hourly cost/billing strategy statistics for nodes where the application runs	Includes billing strategy statistics for nodes where the application runs, and hourly cost and billing strategy statistics for those nodes. This is used to analyze the distribution and trends of different billing strategies for the nodes where the application is running.
Pod resource request ranking Pod resource utilization ranking - Sorted by CPU utilization Pod idle resource ranking	You can analyze departmental cost consumption through cost analysis to help you identify applications with high idle rates. The metrics are as follows: Pod Resource Request Ranking: Statistical analysis of applications with high pod resource requests in the cluster, suitable for scenarios such as capacity planning. Pod Resource Utilization Ranking: You can view pod application replicas with low resource utilization to help discover wasteful applications. Pod Idle Resource Ranking: This metric indicates the proportion of a pod's unused resources to its allocated resources, which reflects the pod's resource waste. By identifying the pods with the most resource waste in a namespace, you can directly identify the main applications that cause resource waste in the namespace. This helps you analyze the reasons for the waste and design targeted resource optimization strategies.

FAQ

Why is no data displayed after I enable cost visualization?

You can check whether a NAT Gateway is configured for the cluster. Some regions do not support querying bills through internal endpoints. Make sure that your cluster can access the Internet.
The Day-over-day cost change and Tomorrow's predicted cost metrics are displayed only after cost data has been collected for two consecutive days.

Why does the sum of namespace costs not equal the actual bill?

Namespace costs are calculated based on cost estimation, not directly from bill analysis. Therefore, cost estimation is based on the list price. When the cluster cost includes deductions from coupons, discounts, or savings plans, data drift may occur. However, you can allocate costs by multiplying the total cluster cost by the cost percentage of each namespace.

Why don't the bills show all the cloud products that the cluster uses?

Cost analysis includes statistics only for cloud products that are exclusively used by the current cluster. Cloud products shared by multiple clusters are not included in the bill statistics for cost analysis.

The Cost Insights feature relies on the cost allocation tags feature in the Expenses and Costs console. It uses a specific tag (key:value=ack.aliyun.com:{{ClusterId}}) to track and collect cluster costs. If you disable this tag on the Cost Allocation Tags page, cluster-level cost statistics will become unavailable. To resolve this issue, you must re-enable the ack.aliyun.com and ack.alibabacloud.com/nodepool-id tags.

Why is the cumulative cost for this month or week lower than the actual cost?

The system starts to collect cost data only after the Cost Insights component is enabled for the cluster. Data generated before the component was enabled is not included.