TencentDB for SQL Server Strengths
Official License for Security Compliance
It supports all SQL Server editions from 2012 to 2025, meeting the needs of both legacy systems and modern requirements. With genuine Microsoft licensing, it provides feature support for the corresponding SQL Server versions. This helps avoid the risks associated with unlicensed software use, making your business more trustworthy in a competitive market. Additionally, it offers a full-stack, in-depth security defense covering disks + TDE + SSL + auditing, building an end-to-end security loop.
Note:
The names and trademarks of SQL Server, Windows, .NET, SQL Server Management Studio (SSMS), and others are owned by Microsoft Corporation or its affiliates.
High Performance and Elastic Scalability
It uses enterprise-grade PCIe SSDs to support high-concurrency, high-intensity commercial workloads. The architecture offers flexible combinations of horizontal single/dual/multi-node setups with diverse specifications. Replica nodes can be directly opened for read-only access. Vertical scaling, including instance specification adjustments and storage scaling, is completed within minutes to meet rapidly changing business demands.
Backup for Stability and Reliability
The system provides a triple backup system combining automatic + manual + cross-region methods, along with second-level log protection and flexible policy configuration. When log backup and retention policies are met, point-in-time recovery is supported based on product capabilities. The specific recovery scope and time required are subject to the product documentation and actual conditions to avoid data loss. The full-link backup policy flexibly adapts to business needs with multiple backup mechanisms. Combined with full-link monitoring and intelligent cleanup mechanisms, it ensures enterprise data security, effectively controls storage costs, and meets stringent disaster recovery requirements.
Multi-Layer Disaster Recovery for Business Continuity
The primary/secondary two-node architecture and multi-node HA architecture, based on self-developed keepalive + smart primary node election mechanisms, enable rapid failover in case of failures to ensure business continuity. Service availability reaches 99.99%. Intra-city multi-AZ deployment and cross-region backup for disaster recovery are supported. For high availability, management and control HA uses a 3-AZ setup across intra-city + cross-city locations. Full-linkage redundancy design from database instances to backend control systems ensures stable business operation.
Smart Ops and Convenient Management
Users do not need to worry about database installation and maintenance. They can manage instances, permissions, networks, logs, backups, and monitoring through the Tencent Cloud console and SQL Server Management Studio (SSMS). Ops becomes less complex. Nearly 50 real-time metrics and custom thresholds for alarms are supported to quickly locate issues.
Business Intelligence for Deep Empowerment
The built-in SSIS + BI full-linkage analysis service covers data extraction, ETL, and visual analysis and helps enterprises explore high-value data, perform business intelligence analysis, and build a primary data management system.
High Availability/Disaster Recovery Capabilities of TencentDB for SQL Server
TencentDB for SQL Server provides disaster recovery capabilities at various levels, including instance, server, data center, AZ, and region. When disasters such as natural disasters, equipment failures, or human operational errors occur, these capabilities ensure stable operation of the production system while minimizing data loss. Backup is the cornerstone of all disaster recovery efforts and the last line of defense for data high availability. TencentDB for SQL Server also offers comprehensive backup capabilities. Their purpose is to enable data recovery in the event of a complete system data crash, striving to ensure business recoverability with minimal time consumption.
TencentDB for SQL Server provides instances with diverse architectures to ensure high availability:
Single-node instance (formerly Basic Edition)
The architecture is deployed on CVM with storage-computing separation. The underlying Premium Cloud Disk three-replica storage can prevent data loss. In extreme cases, if an instance fails, a new instance will be started. The data in the original instance is automatically restored to the new instance through backup recovery by combining data backup and the most recent log backup. The specific recovery time depends on the size of the data volume. Typically, the servers of the two database instances are on the same physical server.
Two-node instance (formerly HA Edition and Cluster Edition)
The primary-secondary two-node architecture for SQL Server 2012, 2014, and 2016 Enterprise consists of a one-primary-one-mirror SQL Server database. It is deployed across racks/AZs and provides second-level HA automatic failover.
The primary-secondary two-node architecture for SQL Server 2017, 2019, 2022, and 2025 Enterprise uses Always On technology to build a high-performance, highly available, highly reliable, and easy-to-maintain SQL Server cluster architecture across racks/AZs. This provides second-level HA automatic failover.
Multi-node instance
The one-primary-multiple-replica architecture for SQL Server 2017, 2019, 2022, and 2025 Enterprise uses Always On technology to build a high-performance, highly available, highly reliable, and easy-to-maintain SQL Server cluster architecture across racks/AZs. This provides second-level HA automatic failover.
For intra-city disaster recovery, multi-AZ deployment capability is provided. Private networks between AZs within the same region are interconnected, and fault isolation can be achieved between AZs. For primary/secondary two-node instances, the primary and secondary nodes can be distributed across different AZs within the same region, with one primary node in the primary AZ and one secondary node in the secondary AZ. This enhances business continuity to prevent database instance failures. Meanwhile, the console supports self-service failover to verify business robustness. Failover within an AZ or between AZs is transparent to applications.
For cross-region disaster recovery, cross-region backup capability is provided. This capability supports storing backup files in another region for storage, setting cross-region backup retention periods, and configuring multiple backup regions simultaneously. After you enable the cross-region backup policy, the instance's backup files are automatically backed up to the target region. If an instance in one region fails, you can use the backup files to restore a new instance in a different region to recover your business. You can also build a cross-region cluster using cross-region read-only instances. Data is synchronized between the primary instance and the read-only instances. If the region of the primary instance fails, you can switch your business to a read-only instance in a different region. This enables fast cross-region failover. The high availability of database recovery provides technical support for customers to implement data availability construction, cross-region backup/recovery, cross-region disaster recovery, long-term archiving, and related compliance management.
TencentDB for SQL Server also provides comprehensive backup capabilities to ensure data security and prevent data loss or corruption. These capabilities include automatic backup, manual backup, data backup, log backup, single-database backup, packaged backup, instance backup, multi-database backup, custom backup policies, customizable backup retention periods from 3 to 7300 days, and customizable backup cycles.
In addition to building comprehensive disaster recovery capabilities at the data layer, we have also established robust disaster recovery capabilities at the management and control layer. Cross-region disaster recovery for databases is only valuable when accompanied by cross-region disaster recovery for the business. However, cross-region distributed deployment of the business inevitably introduces the split-brain problem. At the business management and control layer, our HA is deployed across 3 AZs, with 2 AZs in the same city and 1 AZ in a different city, to ensure stable business operation. Furthermore, during an actual failover, the synchronization status of the database (via database synchronization system tables) is checked to ensure it is normal, thereby preventing erroneous failovers.