tencent cloud

TencentDB for SQL Server

Usage Instructions

Download
Focus Mode
Font Size
Last updated: 2026-06-18 10:53:57

Feasibility of Data Synchronization

Category
Description
Source/Target database type
Source database type:
Self-built databases (IDC self-built databases and self-built databases on CVM): SQL Server 2008 R2, 2012, 2014, 2016, 2017, 2019, and 2022.
Note:
If you need to migrate self-built databases to the cloud, we recommend using the DTS migration feature for greater stability. For operations, see Migration Operation Guide.
Third-party vendor SQL Server: 2008 R2, 2012, 2014, 2016, 2017, 2019, and 2022.
TencentDB for SQL Server (same-account and cross-account): SQL Server 2008 R2, 2012, 2014, 2016, 2017, 2019, and 2022.
Target database type:
TencentDB for SQL Server (same-account and cross-account): SQL Server 2008 R2, 2012, 2014, 2016, 2017, 2019, and 2022.
Synchronization type
Synchronization of structures and full and incremental data is supported.
Synchronization object
Synchronization of database-level data, table-level (user table) data, database and table structures, primary keys, and indexes is supported.
Synchronization of views, storage process, custom functions, custom data types, plan guides, and architectures is supported.
Synchronization of triggers is not supported.
Data type
Supported data types:
See the Supported Data Types List in this document.
Unsupported data types:
Special types: sql_variant, hierarchyid, geometry, geography.
Impact on the source database
When Data Transfer Service (DTS) is used to perform data synchronization, it occupies certain resources of the source instance, which may cause the source instance load to increase and add pressure to the database. If your database configuration is low, it is recommended to perform synchronization during off-peak hours.
Impacts related to enabling CDC:
A brief table lock is caused by enabling CDC since the creation of CDC-related tables and SQL Agent jobs results in certain resource consumption.
When CDC is enabled for a long time, note that the incremental data stored by the CDC component is retained for 3 days by default, which may consume a large amount of storage space. It is recommended to use the command exec console.sys.sp_cdc_change_job @job_type = 'cleanup', @retention= <time>; to set the retention time.
Enabling CDC for a long time can also impact transaction log truncation.
Due to CDC limitations, if the number of tables exceeds 1000 and the number of databases exceeds 10, it is recommended to use the data migration or backup and recovery capability.
The default maximum size for a single field that CDC jobs can handle is 64 KB. If any table with CDC enabled contains a single field larger than 64 KB, the command exec sp_configure 'max text repl size', -1; must be used in advance to adjust the source database configurations.
Impact on the target database
It is recommended not to perform data write operations on the target database when a data synchronization task is executed, as this may affect the progress of the synchronization task and impose the risk of data inconsistency.
Permission
Source: For specific cloud environment and permission requirements, see Multi-Cloud Environment Support.
Target: dbcreator.
If both the source and target are Tencent Cloud SQL Server instances, you can ignore this permission requirement.
Supported SQL statements
DML: INSERT, UPDATE, and DELETE.
Limitation
The data synchronization feature is not suitable for scenarios with a large volume of DML operations within a short period due to the latency.
Foreign key migration is not supported. If tables contain foreign keys, they will not be migrated to the target instance. It is recommended to recreate foreign keys after migration.
DDL operations are not supported. Modifying table structures or deleting tables may cause the synchronization task to fail.
Structure migration cannot be performed independently, and structure and data are required to be migrated together.
Do not modify or delete user information (including usernames, passwords, and permissions) and port numbers in the source and target databases during synchronization, as this may cause the synchronization task to fail.
If the source table is large with long-running transactions during the full synchronization phase, it is recommended to enable READ_COMMITTED_SNAPSHOT.
Instances with SSL encryption enabled cannot be used as source or target instances for data synchronization.

Supported Data Types

Integer Types

Data type
Description
Full synchronization
Incremental synchronization
tinyint
1-byte integer (0 - 255)
smallint
2-byte integer (-32,768 to 32,767)
int
4-byte integer (-231 to 231-1)
bigint
8-byte integer (-263 to 263-1)

Exact Numeric Types

Data type
Description
Full synchronization
Incremental synchronization
decimal(p,s)
Exact decimal
numeric(p,s)
Exact decimal (same as decimal)
money
Currency type (8 bytes)
smallmoney
Currency type (4 bytes)
bit
Boolean type (0/1/NULL)

Approximate Numeric Types

Data type
Description
Full synchronization
Incremental synchronization
float(n)
floating-point number
real
single-precision floating-point number

String Type

Data type
Description
Full synchronization
Incremental synchronization
char(n)
Fixed-length non-Unicode string
varchar(n)
Variable-length non-Unicode string
varchar(max)
Large text (up to 2 GB)
Note:
The default maximum size for a single field that CDC jobs can handle is 64 KB. If any table with CDC enabled contains a single field larger than 64 KB, the command exec sp_configure 'max text repl size', -1; must be used in advance to adjust the source database configurations.
text
Large text
nchar(n)
Fixed-length Unicode string
nvarchar(n)
Variable-length Unicode string
nvarchar(max)
Large Unicode text (up to 2 GB)
Note:
The default maximum size for a single field that CDC jobs can handle is 64 KB. If any table with CDC enabled contains a single field larger than 64 KB, the command exec sp_configure 'max text repl size', -1; must be used in advance to adjust the source database configurations.
ntext
Large Unicode text

Binary Type

Data type
Description
Full synchronization
Incremental synchronization
binary(n)
Fixed-length binary
varbinary(n)
Variable-length binary
varbinary(max)
Large binary (up to 2 GB)
Note:
The default maximum size for a single field that CDC jobs can handle is 64 KB. If any table with CDC enabled contains a single field larger than 64 KB, the command exec sp_configure 'max text repl size', -1; must be used in advance to adjust the source database configurations.
image
Large binary
Note:
The default maximum size for a single field that CDC jobs can handle is 64 KB. If any table with CDC enabled contains a single field larger than 64 KB, the command exec sp_configure 'max text repl size', -1; must be used in advance to adjust the source database configurations.

Datetime Type

Data type
Description
Full synchronization
Incremental synchronization
date
Date (0001-01-01 to 9999-12 -31)
time
Time
datetime
Datetime (1753-01-01 to 9999-12-31)
datetime2
High-precision datetime
smalldatetime
Small datetime (1900-01-01 to 2079-06-06)
datetimeoffset
Datetime with time zone
timestamp/ rowversion
Row version control
Skipped with warning
Skipped with warning
Note:
About timestamp: The SQL Server timestamp (that is, rowversion) is an auto-generated row version column that does not allow explicit INSERT/UPDATE operations. During full export and incremental synchronization, DTS automatically skips this column, and the target generates new values for it. The timestamp column in the table structure is migrated normally (that is, the target creates the column).

Special Types

Data type
Description
Full synchronization
Incremental synchronization
uniqueidentifier
GUID/UUID
xml
XML data

User-Defined Types

Data type
Description
Full synchronization
Incremental synchronization
User-defined Data Type
Alias type based on system types

CDC Support for Multi-Cloud Environments

Use Cases
Migrate from third-party cloud vendors.
Feature Description
Automatically identify the cloud environment.
Enable stored procedures using the corresponding cloud provider's CDC.
This feature reduces permission requirements. In some cloud environments, sysadmin permissions are not required.
Supported Cloud Environments
Cloud Environment
CDC Enablement Method
Minimum Permission Requirement
Self-built SQL Server
EXEC sys.sp_cdc_enable_db
If CDC is not enabled, the minimum permission requirement is sysadmin.
If CDC is enabled, the minimum permission requirement is db_owner.
Third-party cloud vendor
EXEC msdb.dbo.rds_cdc_db
db_owner
EXEC sp_rds_cdc_enable_db
db_owner

Supported Capabilities

Feature Category
Sub-feature or Description
Supported Capability
Resumable transfer
-
Resumable transfer is partially supported during the full export (it is not supported for structure export or export of tables without primary keys).
Resumable transfer is supported for full and incremental imports.
Synchronization direction
One-to-one
Supported.
Many-to-one (multiple instances merged to one/multiple tables merged to one)
Not supported.
One-to-many
Not supported.
Bidirectional synchronization
Not supported.
Synchronization within the same instance (only one-to-one mode is supported)
Not supported.

HA Switch

A switch of the source SQL Server will cause the DTS synchronization task to fail.

Business Impact

When DTS is used to perform full data synchronization, it reads all data of the source database once, which increases the load on the source database. If your database specification is low, it is recommended to execute the synchronization task during off-peak hours or reduce the DTS rate before the task is started.
The impact varies depending on the source database specifications. For example, taking the specification of 8 cores and 16 GB as an example, DTS uses 8 concurrent threads (adjustable) by default. Assuming that there is no network bottleneck, the impact of DTS tasks on the performance of the source database is as follows.
During the full export phase: DTS consumes approximately 18% - 45% of the source database's CPU, adds approximately 40 - 60 MB/s of query pressure to the source database, and occupies about 8 active session connections.
During the incremental export phase: DTS exerts minimal pressure on the source database. By default, eight concurrent connections are used to query CDC data simultaneously.

Business Migration After Data Synchronization

DTS provides a data consistency verification feature in the data synchronization feature. It adopts static data consistency verification to help users verify data consistency for safer service migration. For details, see Data Synchronization Cutover Instructions.

Help and Support

Was this page helpful?

Help us improve! Rate your documentation experience in 5 mins.

Feedback