Updated on 2022-12-16 GMT+08:00

Online Scale-Out

Background

As your business expands, your current system may be unable to provide sufficient disk capacity and satisfactory performance. The distributed GaussDB(DWS) cluster allows you to fully utilize existing machines to scale the existing capacity out, meeting your business needs.

Technical Principles

The node group technology of GaussDB(DWS) supports the scale-out of multiple tables in parallel with a speed up to 400 GB per hour on each new node.

Figure Figure 1 shows the GaussDB(DWS) scale-out process.

Figure 1 Scale-out

Benefits

Online scale-out of GaussDB(DWS) has the following advantages:

  • Scale-out does not interrupt services.

    Data import and query in GaussDB(DWS) are not interrupted during scale-out.

  • Consistent hashing and multi-table parallel scale-out technologies improve query performance.

    Hash technology is used to minimize the amount of data migrating during redistribution.

    Tables can be redistributed in parallel or in user-specified order.

    Users can query the scale-out progress in real time.

  • As the number of nodes increases, the cluster performance improves linearly.

    As shown in Figure 2, in the full parallel distributed architecture, the data loading performance, service processing performance, and capacity of a GaussDB(DWS) cluster linearly improves with the number of nodes.

    Figure 2 High-performance scalability