Overview
If you need an M Node to access the high-performance file systems provided by Cloud Object Storage (COS) or Cloud HDFS (CHDFS) with metadata acceleration enabled, specify specific storage files and mount points in the Elastic MapReduce (EMR) console. The system will automatically generate a CHDFS permission group and bind it to the configured CHDFS storage files and mount points. This way, M Node can access the corresponding file systems and mount points under the COS or CHDFS storage files with metadata acceleration enabled.
The metadata acceleration is a high-performance file system feature provided by COS. Metadata acceleration leverages the powerful metadata management feature of CHDFS at the underlying layer to allow you to access COS over file system semantics. For details, see Metadata Acceleration Overview . Note:
Only existing CHDFS files and mount points can be bound. If you need to create a CHDFS storage file or a mount point, please go to the CHDFS console. Directions
1. Log in to the EMR console, and click the ID or name of the target cluster in the cluster list to go to the Cluster Overview page. 2. On the Cluster Overview page, choose Instance Information > Authorization Information, and click Settings next to M Node Accessing CHDFS.
3. In the Settings pop-up window, you can bind storage files and mount points. You can add, delete, and modify storage files and mount points.
4. First, select a storage file, and then select a mount point. You should select one or more mount points.
Note:
You can select storage files and mount points from multiple buckets.
You can unbind storage files and mount points from system-generated permission groups.