Table Store provides the following benefits:
Dynamic adjustment of reserved read/write throughput
When creating a table, you can configure the reserved read/write throughput for an application based on business requirements and data access conditions. Table Store schedules and reserves resources, based on the table’s reserved read/write throughput, to minimize the resource usage costs. You can dynamically adjust the table’s reserved read/write throughput based on the application.
The amount of data stored in Table Store tables is unlimited. If a table size increases, Table Store adjusts the data partitions for immediate storage space allocation to the resized table.
Table Store stores multiple data copies across different servers in different racks. If a failure occurs, backup servers with copied data immediately restore services, resulting in zero data loss.
Through automatic failure detection and data migration, Table Store protects applications from both hardware and network-related faults to deliver high availability.
Ease of management
Table Store automatically manages complex tasks, such as the management of data partitions, software upgrades, hardware upgrades, configuration updates, and cluster resizing, allowing you to focus on growing your business.
Secure access platform
Table Store performs identity authentication for each application request, preventing unauthorized data access and ensuring data access security.
Table Store guarantees high consistency of writing data. Once a successful result is returned for a write operation, applications can read the latest data.
Flexible data models
Table Store tables do not require a fixed format. The column numbers of each row, and the value types in columns of the same name but different rows, can be varied. Table Store supports multiple data types, such as Integer, Boolean, Double, String, and Binary.
Table Store only charges fees based on the actual resources you have reserved and used.
The Table Store console provides real-time monitoring information, including the requests number per second and the average response latency.