Configuration – scaling
Elastic scale-out is at the core of ShardingSphere, and its relevant configurations are placed in the sharding rule YAML configuration. Although it requires a long process, it can be easily operated through DistSQL.
Scale-out has two modes, namely an automated process and a manual process. The completionDetector configuration should be active in automated processes, while the other configurations can be turned on according to actual needs. Then, trigger elastic scale-out [job] until it is done. In the manual process, each stage of elastic scale-out can be controlled.
This section will discuss two things:
- The elastic scale-out job DistSQL syntax and examples
- The elastic scale-out YAML configuration
DistSQL for job management
Users can control the whole process of scaling data migration through DistSQL, including starting and stopping scaling jobs, viewing progress, disabling reads, checking scaling, switching configurations...
