51Degrees Pipeline Documentation
The Pipeline API has been created by 51Degrees to remove friction from the process of working with real-time data micro-services. It is the mechanism for delivering 51Degrees' data services such as device detection but it is more than that.
The Pipeline aggregates multiple different services from multiple vendors into a single, simple request. The plug-in based architecture makes it easy for new services to be added by third parties. You can also add your own custom elements, taking advantage of the features built into the Pipeline and allowing you to focus on your business logic.
Fundamentals
Each Pipeline is defined by the flow elements that are added to it. Each flow element performs some data processing task.
When a Pipeline is created, the desired flow elements are added to it and can be configured to run in sequence or in parallel as required. (If the language supports parallel execution).
Once created, a Pipeline can then be used to create flow data. This is a one-time-use construct that is used both to pass data to the Pipeline for processing and to access the results.
After the input data (evidence) has been added to the flow data, it is processed by the component flow elements that make up the Pipeline. Each flow element can add new values to the flow data, which can be retrieved once processing is complete.
For a deeper understanding of the Pipeline concept, start with the Pipeline page.
Key features
- Seamless switching between on-premise data and cloud backed data services.
- Automatic updating of on-premise data files from a configurable central service, including in place updates to avoid the need to interrupt running processes.
- For web environments, JavaScript bundling to improve client side performance and network efficiency.
- Support for high performance native data processing libraries via SWIG.
- All components designed for scalable, multi-threaded applications.
- Modern software design including fluent builders, dependency injection, unit testing and continuous integration. DevOps friendly.
- Support for .NET, Java, PHP and Node with more being added over time.
- Plug-in architecture supporting custom third party data services.
New to the Pipeline?
- If you're migrating from the older 51Degrees device detection API then start with the version 3 migration guide.
- New users should check the quick start guide for help getting up and running as soon as possible.
51Degrees Data Services
- Device Detection - Detailed insight into the devices being used to access your content.
- Reverse Geocoding - Obtain accurate real-world data such as postal address from the users' location as reported by their smart phone or similar device.
Technical Detail
- Explore detailed technical documentation for each language supported by the Pipeline.
- Build a custom engine to match your requirements.
- Compare features of Pipeline and data service implementations in each language.