Write Performance Test

YMatrix provides a graphical interface to experience database capabilities. You can customize the data ingestion scale based on business scenarios directly from the web interface and experience YMatrix's powerful data write performance.

Note!
Database performance also depends on server hardware. We recommend using physical machines for testing whenever possible.

The server must meet the following minimum requirements:

CPU Memory Disk Space
8 cores 16 GB 50 GB

1 Preparation

1.1 Log In

Enter the IP address of the machine where MatrixGate is running (by default, the IP of mdw) and the port number in your browser:

http://<IP>:8240

After successful login, you will be directed to the data stream list page.

2 Write Performance Test

Select the "Write Test" module under the "Quick Experience" section.

2.1 Generate Data Stream

Customize the volume and characteristics of data ingestion according to your business scenario. The following four common parameters are available:

  1. Number of Devices: The number of devices reporting data. For example, in a smart factory scenario, this could be the total number of machining tools in a workshop.
  2. Number of Metrics: The number of metrics reported per device. For example, in a smart factory, this refers to all process indicators collected from one machine tool.
  3. Collection Frequency: How often metrics are uploaded. For example, in a smart factory, how frequently temperature data is sent. Use the average collection frequency across all device metrics.
  4. Timestamp Range: The time range for simulated data generation.

After filling in these parameters, the page displays the "Estimated Generated Data". Adjust any of the above four parameters to modify the expected data volume.

To support diverse data characteristics across business scenarios, advanced settings are also provided. Click "Advanced Settings" to configure the following four optional parameters:

  1. Null Rate: Some metric values may be null due to business characteristics. Enter an estimated percentage.
  2. Randomness Level: Describes the fluctuation level of metric values. Default is medium.
  3. Out-of-Order Reporting: Due to network conditions, data may not arrive in chronological order. Estimate and enter the expected out-of-order ratio.
  4. Batch Reporting: In some cases, a device may report its metrics in multiple batches. Enter the average number of batches per report.

2.2 Write Performance

Once the test starts, data begins ingesting into the database. The page updates the data ingestion rate and compression ratio every 5 seconds, clearly showing the database performance. You can stop the ingestion at any time.