Collecting Server Performance Data

Notes

• After a collection starts, performance data is collected every 5 minutes by default.
• The collection duration should be at least 1 hour. If the collection duration is too short, the maximum and average values in 7- and 30-day history cannot be calculated.
• A performance collection runs for seven days by default and stops after the period ends. You can stop or restart the collection at any time. After the collection is restarted, the collection period is recalculated.
• Collected performance data will be retained for 180 days and automatically deleted after that.
• It is recommended that the MgC Agent be installed on an independent server to ensure collection stability and efficiency.
• The MgC Agent can collect performance data from up to 1,000 Linux servers at the same time. If there are 1,000 or more Linux servers to be collected, the server where the MgC Agent is installed must have at least 8 CPUs and 16 GB of memory with at least 8 GB available.
• The MgC Agent can collect performance data from up to 500 Windows servers at the same time. If there are 500 or more Windows servers to be collected, the server where the MgC Agent is installed must have at least 16 CPUs and 32 GB of memory with at least 8 GB available.

Prerequisites

• You have completed a discovery task over the Internet or an intranet, or you have manually added servers to MgC.
• You have installed the MgC Agent in the source intranet environment and have connected the MgC Agent to MgC.
• You have added source server credentials to the MgC Agent.

Procedure

1. Sign in to the MgC console. In the navigation pane, under Project, select your application migration project from the drop-down list.
2. In the navigation pane, choose Source Resources.
3. On the Servers tab, configure MgC agents and credentials for the servers whose performance data needs to be collected.
   • If the Start button in the Performance Collection column is dimmed, move the cursor to the Start button. In the dialog box that is displayed, click Configure Now.

     

   • If the Start button in the Performance Collection column is available, the configuration is complete. In this case, skip this step and go to step 7.

4. Configure the parameters listed in Table 1.

   Table 1 Parameters for configuring migration readiness

   Parameter	Configuration
   Type	Set this parameter based on the source server OS type.
   MgC Agent	Select the MgC Agent installed in the source environment.
   Access IP Address	Select the IP address for accessing the source server. It can be a public or private IP address. If you need to use a proxy to access the server, enter the proxy IP address. After the migration pre-check passes, the IP address will be used for subsequent migration.
   Port	Enter the port on the source server that allows access from the MgC Agent. By default, port 5985 on Windows source servers must be opened to the MgC Agent. The port cannot be changed. By default, port 22 on Linux source servers must be opened to the MgC Agent. You can specify a different port if needed.
   Credential	Select the server credential. If the credential is not displayed in the list, go to the MgC Agent console, add the server credential, and synchronize it to MgC.

5. Click Confirm. The system will verify the correctness of the configuration.
6. After the configuration is verified, click Start in the Performance Collection column to start collecting server performance data. The collection status will change to Collecting.

   After performance data collection starts, you can:

   • View the collected data.
     1. Click a server name to go to its details page. In the Performance Info area, you can view the collected performance data. The maximum values in 7 days and 30 days are hourly aggregated data. To view these maximum values, ensure the collection duration exceeds 1 hour.
     2. Click View in the Operation column to view the details and line chart about the performance metric in the collection period.
   • Stop the collection.

     Click Stop to pause performance data collection.