Products & Solutions / Event Stream Processing

SAS® DataFlux® Event Stream Processing Engine

Gain immediate analytic insights from real-time data streaming into your organization

It's a world of constant information exchange. Every organization has real-time data flowing both from outside and from within. Customer transactions, internal processes, changing market conditions and burgeoning social media add to the volumes of data that organizations must access and deal with each day. Traditional approaches, which apply analytics after data is stored, may provide insights too late.

The SAS DataFlux Event Stream Processing Engine continuously analyzes data as it is received. The software is a form of complex event processing (CEP) technology that is often used for mission-critical data and decision applications. It empowers you to make important decisions in real time.

Real-time decision-making opportunities include:

  • Fraud detection to ensure that every bank transaction is valid.
  • Real-time analysis of social data streams to understand customers' sentiments about your products.
  • Personalized online offers based on consumers' website search criteria.

Benefits

  • Delivers information in real time.
  • Provides increased performance for real-time decision management.
  • Handles complex logic and improves storage latency.

Read more

Features

  • Native C++ object modeling application programming interface (API)
  • Powerful expression language for scripting complex processing logic
  • Seamless interaction with SAS solutions and capabilities
  • Windows for filtering data
  • Procedural and pattern-matching capabilities
  • Ability to pass batches of real-time information (window pulsing) for performance tuning
  • Easily configurable levels of determinism
  • Flexible threading by project

Read more

How SAS® Is Different

  • Deep and proven CEP and event stream processing (ESP) expertise. The SAS DataFlux Event Stream Processing Engine is the third CEP framework from SAS. The SAS DataFlux Event Stream Processing Engine is built with performance considerations that allow real time processing of incoming streams of data.
  • Easy integration with other SAS technologies. Existing SAS customers can easily integrate event stream processing capabilities with other SAS capabilities such as data visualization, reporting and analytics. Activity logs can be used for auditing and performance tuning as desired.
  • Procedural windows are supplied for creating event processing rules. The SAS DataFlux Event Stream Processing Engine enables you to create or recreate business and decision processes using both out-of-the-box and procedural windows to route events for appropriate handling. Complex logic can be easily implemented using C++ scripting in a custom window.
  • Sophisticated pattern matching. A unique pattern-matching facility allows you to define sequential or temporal (time-based) events, which can then be used to monitor breaks in patterns so corrective action can be taken immediately. You do not have to wait for batch jobs to run before receiving critical information.

Benefits

  • Delivers information in real time. Event stream processing allows the continuous analysis of events as they occur. This incremental updating of information enables the real-time analysis of trends to detect anomalies immediately. These capabilities help capture value that could otherwise be lost through information lag.
  • Provides increased performance for real-time decision management. The SAS DataFlux Event Stream Processing Engine has been designed to outperform existing ESP engines by using SAS technologies, such as parallel processing, threaded kernel processing and customizable data filters. It also provides native support for updates, deletions and insertions for easy modeling with improved performance.
  • Handles complex logic and improves storage latency. The ability to split incoming events enables you to conserve bandwidth for better performance as well as generate multiple processing paths for complex events. A powerful expression language and procedural windows provide better methods for handling complex processing logic.

Features

Native C++ object modeling application programming interface (API)
  • Standard, well-known programming language.
  • Custom algorithms that are easy to integrate.
  • Standard development environment, including debuggers.
  • Can be embedded in the application or as a standalone server.
  • Native C++ data structures.
Powerful expression language for scripting complex processing logic
  • Uses the powerful DataFlux® expression engine.
  • Lets you create field calculation expressions and data filtering expressions easily.
  • Performs faster as a result of expression execution overhead.
Seamless interaction with SAS solutions and capabilities
  • Uses the SAS threaded kernel library.
  • Provides platform independence for multiple operating system support.
  • Provides the ability to use SAS Analytics within the event stream processing procedural window.
Windows for filtering data
  • Ability to define splitter functions or expressions to determine the output paths for generated events.
  • Maximizes bandwidth by not sending unnecessary events to additional filter windows.
Procedural and pattern-matching capabilities
  • Enables you to write C++ event stream headers.
  • Enables detection of event patterns of interest over time.
  • Procedural windows enable the scripting of complex logic to be handled easily.
  • Pattern-matching facility allows you to define sequential events that can be used to monitor breaks in patterns so corrective actions can be taken immediately.
Ability to pass batches of real-time information (window pulsing) for performance tuning
  • Supports the ability to forward only new events from a window at a given pulse period.
  • Reduces redundant data updates.
  • Generates snapshots of differences.
Easily configurable levels of determinism
  • Ability to set a single thread pool for high determinism and ease of initial testing.
  • An enhanced tagged-token mechanism is available for high determinism in a multithreaded model.
  • Ability for multithreaded runs without a tagged token for the fastest, most consistent model execution.
Flexible threading by project
  • Enables you to dictate your event stream processing to use parallel processing (i.e., splitting the processing along different cores in your processor) for improved performance.
  • Lets you add tokens as data stream identifiers for higher levels of determinism, which is useful for testing.
  • Uses mixed modes for multiple projects.
  • Allows user-tuning of the project thread pool size for faster identification of anomalies.

System Requirements

Supported  Platforms (minimum versions)
  • HP-UX  Itanium: HP-UX 11.31 (11iv3)
  • Linux 64-bit: RHEL5 and SLES10
  • Solaris on SPARC: Version 10 Update 6
  • Solaris on x64: Version Solaris 10 Update 6
  • Microsoft Windows 64-bit:  Windows XP Professional, Windows Server 2003, Windows Server 2008, Windows Vista*, Windows 7**

* NOTE: Windows Vista supported editions are: Enterprise, Ultimate and Business.
** NOTE: Windows 7 supported editions are: Enterprise, Ultimate and Professional.

Please contact your local SAS sales representative for complete details on configurations and system requirements.

Ready to learn more?

Call us at 1-800-727-0025 (US and Canada) or request more information.