Release notes

Feature

Description

Release date

Available regions

References

DML operations for non-transactional tables

Non-transactional tables now support INSERT, UPDATE, and DELETE statements through SQL.

2026-05-15

All regions

DML operations

Feature

Description

Release date

Available regions

References

Active-active mode for global tables

Active-active global tables let you write data in multiple regions simultaneously. The system aggregates writes into a global dataset and serves eventually consistent reads, supporting cross-region scenarios such as game user profiles and social messaging systems that need shared global data.

2026-01-30

China (Hangzhou), China (Shanghai), China (Qingdao), China (Beijing), China (Zhangjiakou), China (Hohhot), China (Ulanqab), China (Shenzhen), China (Guangzhou), and China (Chengdu)

Global Table

Feature

Description

Release date

Available regions

References

Global tables

Global tables synchronize data across regions automatically, enabling cross-region disaster recovery with low-latency local reads and writes and failover. Use them when cross-region workloads require both low latency and high availability.

2025-12-10

China (Hangzhou), China (Shanghai), China (Qingdao), China (Beijing), China (Zhangjiakou), China (Hohhot), China (Ulanqab), China (Shenzhen), China (Guangzhou), and China (Chengdu)

• Global Table

• Global table limits

• Global table metering and billing

Feature

Description

Release date

Available regions

References

Resource plan upgrade

China-wide resource plans are upgraded to global resource plans. Region-level plans for Hangzhou, Shanghai, Beijing, and Shenzhen, along with the Overseas Region A general resource plan, are discontinued. Active plans are not affected.

2025-08-01

All regions

Resource plans

Feature

Description

Release date

Available regions

References

Custom timeline identifiers and primary-key data fields for the TimeSeries model

• Define custom timeline identifier fields to fit your business model.

• Add data fields as primary keys to store multiple rows that share the same timeline identifier and timestamp but carry different time series data.

2024-06-28

China (Hangzhou), China (Shanghai), China (Beijing), and China (Zhangjiakou)

Specify custom time series identifiers and custom data fields as the primary key columns

Lastpoint index for the TimeSeries model

The Lastpoint index retrieves the most recent point of each timeline in a time series table, helping you fetch the latest data of monitored targets and speed up query and analysis.

2024-06-28

China (Hangzhou), China (Shanghai), China (Beijing), and China (Zhangjiakou)

Lastpoint index

Search index storage price reduction for VCU mode (formerly reserved mode)

The storage price of search indexes in VCU mode (formerly reserved mode) is reduced. For details, see Search index storage price adjustment notice for VCU mode (formerly reserved mode).

2024-06-20

All regions

Product pricing details

Reserved VCU price reduction in VCU mode (formerly reserved mode)

The reserved VCU price is reduced. For details, see Reserved VCU price adjustment notice.

2024-06-01

All regions

Product pricing details

Billing mode renaming

• Reserved mode is renamed VCU mode.

• Pay-as-you-go mode is renamed CU mode.

2024-06-01

All regions

• Billing overview

• VCU mode (formerly Reserved mode)

• CU mode (formerly pay-as-you-go mode)

Elastic capability support for VCU mode (formerly reserved mode)

VCU mode (formerly reserved mode) supports prepurchased reserved VCUs based on resource estimates, with the option to enable elastic capacity and pay for actual compute usage. Stack reserved VCUs with elastic capacity to optimize costs.

Enabling elastic capacity for an instance switches it to serverless mode—pay only for actual compute usage with no upfront resource purchase.

For details, see Product upgrade notice.

2024-06-01

All regions

• Billing overview

• VCU mode (formerly Reserved mode)

• CU mode (formerly pay-as-you-go mode)

Feature

Description

Release date

Available regions

References

Data encryption enhancements

When creating a data table in the console, you can now choose BYOK (bring your own key) as the encryption type alongside the KMS service primary key.

2024-05-30

All regions

• Data table operations

• Data encryption

Feature

Description

Release date

Available regions

References

Analytical Store for time series data is now generally available

Analytical Store provides low-cost storage and fast queries and analysis for time series data, designed for long-term retention and analysis. Key capabilities:

• Create, modify, query, or delete an analytical store.

• Synchronize time series data from a time series table.

• Query and analyze data in the analytical store with SQL.

2023-12-20

China (Hangzhou), China (Shanghai), China (Beijing), and China (Zhangjiakou)

• Analytical Store

• Limits on the TimeSeries model

• Billable items of the TimeSeries model

Instance Policy

Configure Instance Policy on an instance to restrict access by source IP, source network, and TLS version, securing both instance access and the resources it holds.

2023-12-13

All regions except UAE (Dubai)

• Configure an instance policy

• Instance policy

Network ACL

Configure Network ACL on an instance to restrict the network types that can reach it, securing network access.

2023-12-13

All regions

• Network ACL

• Network security management

New search index capability

Added compensation sort: when a sort column is null for a row, the sort can fall back to another column's value.

2023-12-01

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Malaysia (Kuala Lumpur), Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

FieldSort

Feature

Description

Release date

Available regions

References

Zone-redundant storage

Create a zone-redundant instance to store data across data centers. If a data center becomes unavailable due to a network outage, power failure, or disaster, Tablestore continues to serve strongly consistent reads and writes, delivering high availability and disaster recovery.

2023-11-19

• High-performance instances and standard (high-performance storage) instances in China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), and Singapore

• Capacity instances and standard (capacity storage) instances in China (Hong Kong)

• Search indexes in China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Singapore, and Germany (Frankfurt)

ZRS

Data backup enhancements

Cloud Backup operations are integrated into the Tablestore console.

To prevent data loss from accidental deletion or malicious tampering, use Tablestore Backup to back up wide-table data in your instance and restore it when needed.

2023-11-10

China (Hangzhou), China (Shanghai), China (Shenzhen), China (Beijing), China (Zhangjiakou), China (Hohhot), and China (Hong Kong)

• Backup and restore

• Back up data

• Restore data

• Configure alert notification for a backup plan

Feature

Description

Release date

Available regions

References

New region support for search index

Search index is now available in Malaysia (Kuala Lumpur).

2023-10-10

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Malaysia (Kuala Lumpur), Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

• Features and regions

• Search index

• Search index limits

Feature

Description

Release date

Available regions

References

Authorization via Resource Directory Control Policy

Use custom Control Policies to define permission boundaries for resource access. Tablestore supports policies that restrict TLS versions and require non-public-network instances.

2023-09-19

All regions

Use custom access control policies

SQL query enhancements

Python SDK now supports SQL queries.

2023-09-04

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

Python SDK version history

Feature

Description

Release date

Available regions

References

SQL queries on search index virtual columns

Query data through search index virtual columns with SQL.

2023-08-28

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

Search index virtual columns

New search index capabilities

• Java SDK adds date histogram aggregation for search index statistics.

• Go SDK adds histogram and date histogram aggregations for search index statistics.

2023-08-17

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

• Java SDK aggregations

• Go SDK aggregations

Feature

Description

Release date

Available regions

References

SQL queries on nested and array types in search indexes

Query nested and array fields in search indexes with SQL.

2023-06-14

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

• Array type in search indexes

• Search Index Nested Type

TimeSeries model enhancements

Use Realtime Compute for Apache Flink to write Tablestore wide-table data into Tablestore time series tables.

2023-06-09

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

Use Realtime Compute for Apache Flink to process Tablestore data

New region support for TimeSeries model

TimeSeries model is now available in China (Hong Kong), Germany (Frankfurt), US (Virginia), and China (Ulanqab).

2023-06-01

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

Features and regions

Feature

Description

Release date

Available regions

References

SQL queries support JSON functions and JOIN

SQL now supports JSON functions and JOIN.

JOIN lets you combine two or more tables and return rows that meet both the join condition and the query condition.

2023-05-23

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

• JSON functions

• Join

Feature

Description

Release date

Available regions

References

Time series data migration

Migrate data from a time series table to another data table through DataWorks Data Integration.

2023-04-26

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), Germany (Frankfurt), and Singapore

Sync data from a time series table to another time series table

Secondary index enhancements

PHP SDK adds local secondary index support.

2023-04-21

All regions

Create a secondary index

Data import tool support

Sync MySQL and Oracle data to Tablestore through Tapdata Cloud.

2023-04-12

All regions

• Sync MySQL data

• Sync Oracle data

Time Series Writer for TimeSeries model

Use the Time Series Writer to write time series data into Tablestore time series tables. The writer supports multi-table writes, write statistics, row-level callbacks, and custom configuration.

2023-04-03

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), Germany (Frankfurt), and Singapore

Write time series data using Timeseries Writer

Feature

Description

Release date

Available regions

References

SQL queries available in more regions

SQL queries are now supported in China (Hong Kong), Germany (Frankfurt), US (Virginia), and China (Ulanqab).

2022-11-15

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Ulanqab), China (Shenzhen), China (Hong Kong), Germany (Frankfurt), US (Virginia), and Singapore

Features and regions

Feature

Description

Release date

Available regions

References

Search index enhancements

PHP SDK search indexes now support statistical aggregations, concurrent data export, and search index updates.

2022-10-31

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

PHP SDK historical versions

SQL query enhancements

PHP SDK now supports SQL queries.

2022-10-31

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

PHP SDK historical versions

Feature

Description

Release date

Available regions

References

Search index enhancements

• .NET SDK search indexes now support statistical aggregations and concurrent data export.

• Node.js SDK search indexes now support statistical aggregations, concurrent data export, search index updates, and collapse (deduplication).

2022-08-16

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

• Version history of Tablestore SDK for .NET

• Version history of Tablestore SDK for Node.js

SQL query enhancements

• .NET SDK now supports SQL queries.

• Node.js SDK now supports SQL queries.

2022-08-16

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• Version history of Tablestore SDK for .NET

• Version history of Tablestore SDK for Node.js

Feature

Description

Release date

Available regions

References

IoTstore IoT storage solution

Tablestore launches the all-in-one IoTstore solution that delivers storage, query, search, analytics, and synchronization for massive IoT device metadata, message data, time series trajectories, and other data.

2022-07-21

• WideColumn model: All regions

• TimeSeries model: China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• IoT Storage (IoTstore)

• Time series data ingestion

• Device time series data development

• Device metadata ingestion

Search index enhancements

Python SDK search indexes now support virtual columns, the date data type, collapse (deduplication), percentile statistics, and histogram statistics.

2022-07-20

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Python SDK version history

SQL query enhancements

You can now add or delete attribute columns for existing mapping tables when you use SQL queries.

2022-07-01

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

Update the attribute columns of a mapping table

Feature

Description

Release date

Available regions

References

Search indexes support the date data type

By mapping Integer or String data in data tables to the date data type in search indexes, you can use the date data type for better-performing range queries.

2022-05-20

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Date data type

Feature

Description

Release date

Available regions

References

Grafana integration

Grafana is an open source visualization and analytics platform. After Tablestore is integrated with Grafana, you can use Grafana to visualize data stored in Tablestore.

2022-04-28

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

Connect to Grafana

SQL query enhancements

Search indexes can now be mapped to tables, and table-level parameters such as consistency and accuracy can be configured to control index selection.

2022-04-28

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• Create a mapping table for a table

• Create mapping tables for search indexes

• Index selection policy

• How do I select a query method when I use SQL to query data?

SQL query enhancements

You can now access Tablestore through a Go language driver.

2022-04-19

Go driver

Feature

Description

Release date

Available regions

References

TimeSeries model supports the multi-value model

When you use SQL to query time series data, you can manually create a multi-value model mapping for a time series table so that the data can be queried in multi-value model form.

2022-03-25

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

Use SQL to query time series data

Back up and restore Tablestore data with Cloud Backup (public preview)

Cloud Backup periodically backs up data in Tablestore instances and restores it when data is lost or damaged. Cloud Backup supports both full and incremental backups, and uses data redundancy to improve the reliability of the backup vault.

2022-03-18

China (Hangzhou), China (Shanghai), and China (Beijing)

• Backup and restore

• Back up data

• Restore data

SQL query enhancements

Search indexes can now match specified strings to support full-text search.

2022-03-09

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

Full-text search

Feature

Description

Release date

Available regions

References

Realtime Compute for Apache Flink integration

Tablestore now integrates with Realtime Compute for Apache Flink for big data analytics.

2022-02-07

All regions

• Solution background

• Preparations

• Implement the solution

Feature

Description

Release date

Available regions

References

SQL query enhancements

You can now use JDBC with Hibernate or MyBatis to access Tablestore.

2022-01-20

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• Use Hibernate to query data with SQL

• Use MyBatis to query data with SQL

Sync Kafka data to time series tables

You can use the kafka-connect-tablestore package to write data from Kafka into Tablestore time series tables.

2022-01-07

China (Hangzhou), China (Shanghai), and China (Shenzhen)

• Sync Kafka data to a time series table

• Configuration description

Feature

Description

Release date

Available regions

References

Search index TTL

Time To Live (TTL) is a search index attribute that defines how long data is retained. The system automatically removes data that exceeds the retention period, which reduces storage usage and cost.

2021-12-07

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Lifecycle management

Fuzzy tokenization

For WildcardQuery patterns of the form *word*, fuzzy tokenization delivers better-performing fuzzy queries.

2021-12-02

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Token-based wildcard query

Feature

Description

Release date

Available regions

References

JDBC support for SQL queries

Connect to Tablestore through JDBC or an SDK.

2021-11-17

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• Run SQL queries through an SDK

• Use SQL queries with a direct JDBC connection

New CLI release

The Tablestore CLI now supports operations on time series tables and time series data, and supports SQL queries against time series data.

2021-11-11

China (Hangzhou), China (Shanghai), and China (Shenzhen)

• Time series table operations

• Data operations

• SQL query

The Tablestore CLI supports operations on data tables, data, search indexes, and Tunnel Service, and supports SQL queries against the data.

2021-11-11

All regions

• Manage data tables

• Operations on data

• Search index

• Tunnel Service

• SQL query

TimeSeries model public preview

Use the TimeSeries model to store, query, and analyze time series data.

2021-11-10

China (Hangzhou), China (Shanghai), and China (Shenzhen)

• TimeSeries model

• Use the TimeSeries model through the console

• Use the TimeSeries model through the CLI

• Use the TimeSeries model by using Tablestore SDKs

Feature

Description

Release date

Available regions

References

Synchronize MySQL data in real time with DTS

With Data Transmission Service (DTS), you can synchronize a MySQL database (such as self-managed MySQL or ApsaraDB RDS for MySQL) to a Tablestore instance for easy data flow.

2021-10-11

All regions

Synchronize MySQL data using DTS

Feature

Description

Release date

Available regions

References

Synchronize Kafka data to data tables

Use the Tablestore Sink Connector to batch-import data from Apache Kafka into Tablestore data tables.

2021-09-22

All regions

• Sync Kafka data

• Sync Kafka data to a data table

• Configuration description

• Error handling

Synchronize MySQL data with canal

For small- to medium-sized databases or individual developers, use canal to synchronize MySQL data to Tablestore. canal is easy to deploy and operate, which makes it suitable for small- to medium-scale MySQL synchronization.

2021-09-10

All regions

Synchronize MySQL data with canal

Feature

Description

Release date

Available regions

References

SQL query public preview

The SQL query feature lets you run complex queries and efficient analytics against Tablestore data, and provides a unified access interface for multiple data engines.

2021-08-23

China (Hangzhou), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), and Singapore

• SQL query

• SQL support

• Use SQL queries through the console

• Create a mapping table for a table

• Delete mapping tables

• Query information about tables

• Query data

• Aggregate functions

• Query the index information about a table

• List mapping table names

• Index selection policy

• Computing pushdown

Feature

Description

Release date

Available regions

References

Virtual columns for search indexes

Modify the search index schema or create a new search index to query new fields and data types, without changing the storage structure or data in Tablestore.

2021-07-28

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Virtual columns

Percentile aggregation

Percentile aggregation reports the percentile distribution of a data set. For example, during routine system operations you often need to inspect the P25, P50, P90, and P99 latency distribution of incoming requests.

2021-07-23

Percentile statistics

Histogram aggregation

Group query results by a specified value interval. Field values that fall within the same range form one group, and the response returns the value and count for each group.

2021-07-23

Group by histogram

Filter supports regular expression matching

Filters now support regular expression matching, including data type conversion before the filter is applied.

2021-07-09

All regions

Filters

Feature

Description

Release date

Available regions

References

Dynamic schema modification for search indexes

Dynamically modify the schema of a search index. For example, add, update, or delete index columns, or change the routing key of the search index.

2021-06-18

China (Hangzhou), China (Hangzhou Finance Cloud), China (Shanghai), China (Beijing), China (Zhangjiakou), China (Shenzhen), China (Hong Kong), Singapore, Indonesia (Jakarta), Japan (Tokyo), Germany (Frankfurt), UK (London), US (Silicon Valley), US (Virginia), and Philippines (Manila)

Dynamically modify search index schema

Synchronize Tablestore data to MaxCompute

Use the merge_udf.jar package in MaxCompute to convert incremental Tablestore data into full-data format.

2021-06-11

All regions

Convert incremental data from Tablestore to a full data format