About the Streaming ML Plugin
The Streaming ML plugin offers a machine-learning powered data ingestion experience with a range of functions that perform in-flight data analysis. Using real time machine learning, models are constantly updated. Trained models both learn on the stream and score on the stream.
You must install the Streaming ML plugin to access these functions.
Streaming ML functions are orders of magnitude faster than alternative approaches to machine learning. On average, these functions process up to 10K records per second (approximately 150 GB per day). Higher throughput volumes may experience some delay.
What the plugin includes
The Streaming ML plugin offers several streaming functions and two aggregation functions.
All Streaming ML functions are currently in beta.
Streaming functions:
- Adaptive Thresholding
- Apply ML Model
- Datagen
- Drift Detection
- Pairwise Categorial Outlier Detection
- Sentiment Analysis
- Sequential Outlier Detection
- Time Series Decomposition (STL)
Aggregation functions:
How to install the Streaming ML plugin
Use the following directions to install the Streaming ML plugin for your .
Prerequisites
- version 1.2 (and above) on-prem is installed on the machine.
- The machine is configured with SCloud command-line tool.
To learn about SCloud, see Get started with SCloud.
Installation steps
- On the master node of the cluster or on a single instance machine, navigate to the working directory.
cd <dsp-folder-name>
- Run the "deploy-ml-plugin" command to install the plugin. You must run the command with root permission.
./deploy-ml-plugin
- You will be prompted to enter the administrator password for your .
Please enter DSP password: {dsp_password}
- Once you enter the password and installation is complete, a success message appears:
[OK] Splunk Streaming ML Plugin is uploaded successfully!
- Reload the browser page and the Streaming ML functions appear.
Uninstall the plugin
To uninstall the Streaming ML plugin, run the command with a -d
option.
The following example shows the command with a -d
option:
sudo ./deploy-ml-plugin -d
Once the uninstall is complete, a success message appears:
[OK] Plugin deleted: 4b4b921a-5455-4bd0-8878-f0920d48c333.
Troubleshoot lookups to the Splunk Enterprise KV Store | Create custom functions with the SDK |
This documentation applies to the following versions of Splunk® Data Stream Processor: 1.2.0, 1.2.1-patch02, 1.2.1, 1.2.2-patch02, 1.2.4, 1.2.5
Feedback submitted, thanks!