Tablestore - Supports Concurrent Data Export
Apr 01 2020Tablestore
Targeted customers: 1. Customers who need to use search index in geo query scenarios or combine search index with data processing engines such as Spark and Presto. 2. Customers who prioritize data export speeds over the sorting of results. Features released: Tablestore search index provides the ParallelScan operation to support the concurrent data export capability. Compared with the Search operation, ParallelScan keeps all query features but analytical features such as sorting and aggregation. As a result, performance is improved by more than five times. Hundreds of millions of rows of data can be exported within one minute. The export capability is unlimited can be scaled out. When the ParallelScan operation is implemented, a maximum of 2,000 rows can be returned. However, a maximum of 100 rows can be returned when the Search operation is implemented. When you call ParallelScan, multithreading is supported while the high export speed is ensured.