Billable items and pricing of VCU mode-Tablestore(Tablestore)-阿里云帮助中心

Billable items in VCU mode (formerly Reserved mode) include computing power, data storage, cross-region replication traffic, and outbound internet traffic. VCU mode helps you save on computing resource fees. This mode also lets you control overall resource usage by configuring an upper limit for elastic capacity or by disabling elastic capacity. This helps you avoid extra fees from unusual traffic, making it a better choice for cost-sensitive scenarios. This topic describes the pricing, billable items, and frequently asked questions (FAQ) for VCU mode.

Pricing

For more information about pricing, see the Tablestore product page.

Basic billing

Billing components

The following figure shows the billing components of VCU mode (formerly Reserved mode).

Billable items

Billable items in VCU mode (formerly Reserved mode) include computing power, data storage, cross-region replication traffic, and outbound internet traffic. The following table describes these items.

Billable item		Billing method	Description
Computing power		Subscription Pay-as-you-go (The elastic part is billed on a pay-as-you-go basis after you enable elastic capacity.)	All computing resources are defined by Virtual Compute Units (VCUs). One VCU is equivalent to the processing power of a 4-core, 16 GB setup. Operations in Tablestore, such as data reads and writes, index writes, queries, analysis, and Tunnel Service, consume VCU computing resources. Important SQL queries do not incur extra fees. However, operations involved in the query process, such as table scans and index lookups, consume VCU computing resources. When you create an instance, you can purchase reserved VCUs based on your required computing resources and pay for them upfront. Tablestore provides a fixed amount of computing power based on the reserved VCUs you purchase. You can also enable elastic capacity for an instance to pay for computing resources based on actual usage. This removes the need to purchase computing resources in advance. After you enable elastic capacity, the elastic upper limit of the instance is the sum of the elastic capacity and reserved VCUs. If the actual computing resource consumption does not exceed the resources provided by your reserved VCUs, Tablestore does not charge extra computing fees. If the consumption exceeds the resources provided by your reserved VCUs, Tablestore charges you for the elastic portion of the VCUs that you use.
Data storage Note If you have a large data volume, use resource plans to reduce storage costs. For more information, see Resource plans.	High-performance Storage	Pay-as-you-go Resource plan	Suitable for latency-sensitive online services that require millisecond-level latency for a single request. Storage scales dynamically based on actual usage and is billed hourly in GB.
	Capacity Storage	Pay-as-you-go Resource plan	Suitable for scenarios with large storage requirements that are not sensitive to read latency. Storage scales dynamically based on actual usage and is billed hourly in GB.
	Search Index storage	Pay-as-you-go Resource plan	When you use a search index, this is the storage space occupied by the index data that Tablestore builds.
Cross-region replication traffic		Pay-as-you-go	When you use global tables, cross-region data transmission traffic is generated during asynchronous data replication to each replica table. Billing is based on the actual amount of synchronized data, in GB. Important The cross-region replication traffic fee for each replica is calculated separately. The fee is charged to the instance from which data is pulled.
Outbound traffic over the internet		Pay-as-you-go	This is the fee for outbound traffic generated when your application accesses Tablestore over the internet. It mainly consists of responses returned from Tablestore over HTTP. Formula: Outbound traffic fee = Outbound traffic (GB) × Price per GB. Note Tablestore charges only for outbound traffic over the internet. Inbound traffic and traffic over the internal network are free. When an operation fails, Tablestore returns a failure message, which also generates outbound traffic. Access between different regions is also considered access over the internet.

Core feature billing

Using core features such as search index, secondary index, SQL query, time series model, multiple versions, lifecycle management, Tunnel Service, global tables, and data migration and synchronization incurs additional fees.

Important

If you access Tablestore over the internet, you will incur outbound traffic fees.
SQL queries do not incur extra fees. However, operations during the query process, such as table scans and index lookups, do incur fees.

The following table describes the billing principles for each feature.

Core feature	Billing principle
Search index	Regardless of the instance type, using a search index consumes High-performance Storage, and index queries consume VCU computing resources. Billable items include High-performance Storage and computing power. For more information, see Search index metering and billing.
SQL query	SQL queries do not incur extra computing costs. However, operations during the query process, such as table scans and index lookups, consume VCU computing resources. Billable items include computing power. For more information, see SQL query metering and billing.
Secondary index	Billable items include index table data storage and computing power. For more information, see Secondary index metering and billing.
Time series model	Billable items include data storage and computing power. For more information, see Time series model metering and billing.
Multiple data versions	This feature mainly incurs extra storage fees for version numbers and the data of each version. For more information, see Data storage.
Lifecycle management	Configuring lifecycle management to delete data does not incur extra fees. However, after you enable lifecycle management, a timestamp is added to each attribute column as a version number, which consumes extra storage. For more information, see Data storage.
Tunnel Service	Tunnel Service itself does not incur extra costs. When you consume data from Tunnel Service, VCU computing resources are consumed based on the amount of data that is pulled.
Global tables	Using global tables consumes data storage and generates cross-region replication traffic for data synchronization between replicas. Data reads and writes also consume VCU computing resources. For more information, see Global table metering and billing.
Data synchronization and migration (using tools or other products)	When you use migration tools or other products, such as DTS or IoT Platform, to migrate data to Tablestore, VCU computing resources are consumed based on the specific read and write requests.
Access from compute engines (MaxCompute, Spark, Flink, etc.)	When compute engines, such as MaxCompute, Spark, or Flink, access Tablestore, VCU computing resources are consumed based on the specific read and write requests.

Resource assessment and selection

When you use VCU mode, you must select a storage class beforehand. For computing resources, you can purchase reserved VCUs based on your business needs, or set reserved VCUs to 0 and enable elastic capacity to achieve a fully pay-as-you-go model. Resource assessment and selection can help users with stable service traffic reduce computing resource costs by planning and purchasing reserved VCUs.

Computing resource assessment

In VCU mode, all computing resources are measured in VCUs. One VCU is equivalent to the computing power of a 4-core, 16 GB configuration. Tablestore operations, such as data reads and writes, index writes, queries, analysis, and Tunnel Service, consume VCU computing resources.

The following table provides recommendations for estimating the required VCUs based on the number of rows read or written per second. Because computing resource consumption is affected by many factors, you must evaluate the required VCUs based on your actual business needs.

Important

The size of rows, request cache hit rate, filter complexity, and search index query analysis complexity all affect the actual VCU consumption. The actual resource consumption depends on your business operations.
To handle sudden traffic peaks and ensure online service stability, we recommend that you reserve about 40% of your computing resources. You can configure the elastic upper limit of the instance to reserve computing resources.
You can enable elastic capacity for an instance to use a serverless model and pay for computing performance based on actual usage. You can also use reserved VCUs at the same time to save costs.

Rows read/written per second	Recommended VCUs	Description
0–1,000	2 VCU	Recommended as the minimum VCU configuration for a production environment. Suitable for access under low loads.
1,000–20,000	4–6 VCU	Suitable for configuration under low to medium loads.
Over 20,000	10+ VCU	For high loads, estimate approximately 5 VCUs for every 10,000 rows read or written per second.

Storage class selection

When you create an instance in VCU mode, you can select either High-performance Storage or Capacity Storage as the instance's storage class. Regardless of the instance's storage class, if you use the search index feature, Search Index storage is used.

Storage class	Performance metrics		Scenarios
Storage class	Avg. write RT	Avg. read RT	Scenarios
High-performance Storage (SSD)	<10 ms	<10 ms	Suitable for latency-sensitive online services, such as advertising, recommendations, and feed streams. It has a low glitch rate, and latency is usually at the millisecond level.
Storage for Search Indexes (SSD)	<10 ms	<10 ms	Regardless of the instance's storage class, this is the storage space occupied by the index data that Tablestore builds when you use the search index feature.
Capacity Storage (SATA)	<10 ms	<100 ms	Suitable for the following scenarios: Online services with general latency requirements, especially those not sensitive to read latency. Scenarios with large data volumes that are cost-sensitive.

Upgrade and downgrade

You can upgrade or downgrade reserved VCUs. Adjust the number of reserved VCUs as needed.

Upgrade operation

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, find the target instance and click Upgrade in the Actions column.

You can also click the instance name. On the Instance Details tab, click Upgrade in the Subscription Information section.
On the Tablestore Standard Instance | Upgrade/Downgrade page, change the Reserved VCU value, read and agree to the terms of service, confirm the price, and then click Buy Now.
On the payment page, check the order and complete the payment. On the payment completion page, you can click Management Console to go to the Tablestore console and view the information about the upgraded instance.

Note

If elastic capacity is enabled and the number of reserved VCUs after the upgrade is less than the current elastic upper limit of the instance, the upper limit remains unchanged. If the number of reserved VCUs after the upgrade is greater than or equal to the current elastic upper limit, Tablestore automatically adjusts the upper limit. The adjusted value is displayed in the console. You can readjust the elastic upper limit of the instance after the upgrade is complete.

Downgrade operation

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, click in the Actions column of the target instance, and then select Downgrade.

You can also click the instance name. On the Instance Details tab, click Downgrade in the Subscription Information section.
On the Tablestore Standard Instance | Downgrade page, change the Reserved VCU value, read and agree to the terms of service, confirm the price, and then click Buy Now.

Note
When you adjust the number of reserved VCUs to 0, you must set Elastic Capacity to Enabled.
Downgrading an instance is free of charge. On the payment completion page, you can click Management Console to go to the Tablestore console and view the information about the downgraded instance.

FAQ

What are the differences between VCU mode (formerly Reserved mode) and CU mode (formerly Pay-As-You-Go mode)?

The main differences between VCU mode and CU mode are as follows:

Computing

In VCU mode, computing power is billed based on a subscription for reserved VCUs (4-core, 16 GB) or by converting actual consumption into VCUs when elastic capacity is used. In CU mode, billing is based on converting actual consumption into CUs.
Storage

In VCU mode, charges are based on different storage classes. In CU mode, storage is tied to the instance type.
Index

In VCU mode, the storage component of an index is billed separately, and the computing component is consumed from the overall computing resources. In CU mode, an index is billed based on High-performance Storage and reserved CUs.

For more information, see Differences between VCU mode and CU mode.

Can I use resource plans to offset data storage fees?

Yes, you can. You can purchase Capacity Storage plans and High-performance Storage plans to offset storage fees. For Search Index storage, you can purchase a High-performance Storage plan to offset storage fees at a 3:1 ratio because of the price difference.

My instance is about to expire. How do I renew it?

You can renew your instance before it expires. The steps are as follows:

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, find the target instance and click Renew in the Actions column.

Note
You can also click the instance name. On the Instance Details tab, click Renew next to Actions in the Subscription Information section to go to the configuration page.
On the Tablestore Standard Instance | Renewal page, select a subscription duration, read and agree to the terms of service by selecting I have read and agree to the Tablestore Standard Instance Terms of Service, and then click Buy Now.
Complete the payment as prompted to renew the instance.

How do I enable elastic capacity for an instance?

If you did not enable elastic capacity when you created the instance, you can enable it later to pay for computing resources based on actual usage. After you enable elastic capacity, the maximum number of VCUs that can be used is determined by the elastic upper limit of the instance. The value of the elastic upper limit ranges from (Number of reserved VCUs + 0.1) to 2000.

Note

After you enable elastic capacity for an instance, you can disable it by clicking Disable Elastic Capacity next to Elastic Capacity in the Instance Basic Information section of the Instance Details tab. On the Tablestore Standard Instance | Upgrade/Downgrade page, set Elastic Capacity to Disabled and complete the subsequent steps as prompted.

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, click the target instance name.
In the Instance Basic Information section of the Instance Details tab, click Enable Elastic Capacity next to Elastic Capacity.
On the Tablestore Standard Instance | Upgrade/Downgrade page, set Elastic Capacity to Enabled, read and agree to the terms of service by selecting I have read and agree to the Tablestore Standard Instance Terms of Service, and then click Buy Now.
Complete the payment as prompted to enable elastic capacity for the instance.

How do I adjust the elastic upper limit of an instance?

After you enable elastic capacity for an instance, the default elastic upper limit is the sum of the elastic capacity and reserved VCUs. If your service traffic increases and you need more computing resources, you can increase the elastic upper limit. The adjustment takes effect immediately. The steps are as follows:

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, click the target instance name.
In the Subscription Information section of the Instance Details tab, click Set Elastic Upper Limit next to Instance Elastic Upper Limit.
In the Instance Elastic Upper Limit dialog box, change the value of the elastic upper limit, and then click OK.

The value of the elastic upper limit ranges from (Number of reserved VCUs + 0.1) to 2000 VCUs. You can upgrade or downgrade the number of reserved VCUs in increments of 1. The elastic upper limit of the instance can be dynamically adjusted in minimum increments of 0.1.

Why does the instance still show the old specifications in the console after an upgrade or downgrade?

The instance upgrade or downgrade is in progress. Wait a few seconds and then check the instance information again.

How do I monitor the VCU usage of an instance?

You can view VCU usage details in the Tablestore console, such as VCU usage, instance specifications, the elastic upper limit of the instance, and elastic capacity usage. The steps are as follows:

Log on to the Tablestore console.
In the top navigation bar, select a region.
On the Overview page, click the target instance name or click Manage Instance in the Actions column.
In the Instance Specifications Monitoring section of the Instance Details tab, click the icon to view the details of the corresponding VCU metrics.