After you create a cluster, you can mount an Apsara File Storage NAS file system on the cluster to improve the shared storage capability of the cluster. This topic describes how to mount a NAS file system and unmount it from a cluster in the Elastic High Performance Computing (E-HPC) console.
The Apsara File Storage NAS service is activated. A NAS file system and a mount target are created. The NAS file system must reside in the same VPC as the cluster. For more information, see Create a file system and Manage mount targets.
The image of the cluster is a public image or a shared image.Important
If a file system is mounted on a cluster that uses a custom image, the file system does not run as expected.
NAS is a distributed file system that provides multiple benefits, including shared access, auto scaling, high reliability, and high performance. To improve the throughput capacity, performance, and concurrent access capability of a cluster, you can mount a file system on the cluster. This way, you can perform high-performance computing on a larger scale.
Mount a file system on the cluster
You can mount the file system on the cluster when you create the cluster. For more information, see Create a cluster. You can also perform the following steps to mount a file system on an existing cluster.
Log on to the E-HPC console.
In the top navigation bar, select a region.
In the left-side navigation pane, choose Resource Management > Storage.
On the Shared Storage page, select the cluster and click the Resource Management tab.
On the Resource Management tab, click Mount Storage Resource.
In the Mount Storage Resource panel, configure the mount settings.
The on-premises directory of the cluster. The user data, scheduler information, and job data of the cluster are stored in the directory.
If no on-premises directory exists, E-HPC automatically creates one.
Storage Instance ID
The ID of the file system.
Mount Point Address
The IP address of the mount target.
Remote Mounting Directory
The remote directory of the file system. E-HPC stores all user data, scheduler data, and job data on a file system. All nodes in the cluster can access the data. This way, the shared storage capability of a cluster is improved.
If no remote directory exists, E-HPC automatically creates one.
After a file system is mounted on a cluster, the file system takes effect for the compute nodes that are newly added to the cluster. However, the existing compute nodes must be reset to apply changes.
Check the mount result
The next time you add or reset compute nodes in the cluster, the file system is automatically mounted on the compute nodes. To check the mount result, perform the following steps:
In the left-side navigation pane, click Cluster.
On the Cluster page, find the cluster and click Connect.
In the Connect panel, specify a username, password, and port number.
Click Browse to view the on-premises directories of the file systems that are mounted on the cluster.
In this topic, an on-premises directory named test is added. The following figure shows the mount result.
Unmount a file system from the cluster
You can unmount file systems from your cluster based on your needs.
On the Storage Management page, select the cluster from the Cluster drop-down list.
Select the file system that you want to unmount, and click Dismount Storage Resource.Note
The file systems in the /ehpcdata, /home, and /opt directories cannot be unmounted.
In the Dismount Storage Resource message, click OK.