Improving HBase Continuous Put Performance
Scenario
In the scenario where a large number of requests are continuously put, setting the following two parameters to false can greatly improve the Put performance.
- hbase.regionserver.wal.durable.sync
- hbase.regionserver.hfile.durable.sync
When the performance is improved, there is a low probability that data is lost if three DataNodes are faulty at the same time. Exercise caution when configuring the parameters in scenarios that have high requirements on data reliability.
This section applies to MRS 3.x and later versions.
Procedure
Navigation path for setting parameters:
On FusionInsight Manager, choose Cluster > Name of the desired cluster > Services > HBase > Configurations > All Configurations. Enter the parameter name in the search box, and change the value.
Parameter |
Description |
Value |
---|---|---|
hbase.wal.hsync |
Specifies whether to enable WAL file durability to make the WAL data persistence on disks. If this parameter is set to true, the performance is affected because each WAL file is synchronized to the disk by the Hadoop fsync. |
false |
hbase.hfile.hsync |
Specifies whether to enable the HFile durability to make data persistence on disks. If this parameter is set to true, the performance is affected because each Hfile file is synchronized to the disk by the Hadoop fsync. |
false |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot