Generating Network Performance Statistics for HyperScale X Nodes

Network performance statistics on the storage/private network for Commvault HyperScale X nodes in a storage pool can be collected using the cv_net_perf.py script. The script calculates network performance statistics between pairs of storage pool nodes, where one node is configured as a server and another as a client.

The script must be executed from one of the nodes in the storage pool. It automatically identifies the other nodes in the storage pool, forms node pairs, and runs the tool sequentially on each pair. The standard data transfer size is 20GB.

Before You Begin

To obtain accurate results, run the network performance tool when there are no backup jobs running against the storage pool.

Procedure

1. Log on to any one of the nodes in the storage pool and navigate to the following directory:

   # cd /opt/commvault/MediaAgent
   

2. Run the following command:

   # ./python3 cv_net_perf.py option

   <option> can be one of the following:

   • CvNetworkTestTool:

     Runs the command using the Commvault network performance tool.

     Example:

     # ./python3 cv_net_perf.py CvNetworkTestTool
     

   • iperf3:

     Runs the tool using the Linux native iperf3 command.

     Example:

     # ./python3 cv_net_perf.py iperf3
     

     Caution

     Do not run the CvNetworkTestTool or iperf3 commands independently on the nodes while the script is running. Any existing instances of CvNetworkTestTool or iperf3 are automatically terminated.

Result

The run time may vary from a few minutes to several hours, depending on the transfer size, workload, and available resources on the nodes.

• The output is logged in the following directory:

  /var/log/commvault/Log_Files/CVFSPerfMon/

• The output is saved using the following naming format:

  network_performance_<timestamp>.log

  Example:

  /var/log/commvault/Log_Files/CVFSPerfMon/network_performance_20221007081957.log

• The SUMMARY section of the log provides a brief overview of network throughput between nodes. The remainder of the log contains detailed run parameters and debug traces.