The following details Table Store’s table, data, and writing operations.
|ListTable||List all tables in the current instance.|
|CreateTable||Create a table.|
|DeleteTable||Delete a table.|
|DescribeTable||Obtain the attribute information of a desired table.|
|UpdateTable||Update the reserved read/write throughput of the table.|
For more details, see Tables of Table Store.
Data operations of Table Store are classified into the following three types:
|GetRow||Read the data of a single row.|
|PutRow||Insert a new row. If a row already exists, the existing row will be deleted first, and the new row will be written in its place.|
|UpdateRow||Update a row. This operation can add and delete the Attribute columns in the row, as well as update the value of the Attribute columns that already exist. If no row is available, this operation will add a new row.|
|DeleteRow||Delete the specified row.|
Batch operation of multiple GetRow operations, which will read several rows of data from one or more tables.
Batch operation of multiple PutRow, UpdateRow and DeleteRow operations, which will insert, update and delete several rows of data from one or more tables.
Read data within a continuous range of the primary key columns in the table.
Table Store writing operations support the following features:
The results of PutRow, UpdateRow and DeleteRow operations guarantee atomicity, as the operations will either succeed or fail completely without any intermediate state.
When a data writing operation receives a successful response, all copies of the data in the distributed file system have been already updated. Any reading operation will then fetch the latest data of the written row.
Furthermore, Table Store provides the operation BatchWriteRow to batch process the write operations of several single rows. The operation can contain all types of the single operation, including PutRow, UpdateRow and DeleteRow. For more information, see Data operations of Table Store.
Note: BatchWriteRow operation does not guarantee atomicity as the response of one batch writing operation may aggregate several success or failure responses for each single writing operation. However, the sub-actions of BatchWriteRow are atomic.