Real-time insight across the board

FeedZai Pulse is a revolutionary next-generation real-time business intelligence platform.

FeedZai Pulse

FeedZai Pulse is a mature enterprise-level data processing engine, based on open standards like Java Enterprise Edition, Java Messaging System, JMX, SQL, HTML5, among others. 
 

A powerful and highly optimized data processing kernel (PKernel) allows hundreds of thousand of events to be processed in real-time providing up-to-date information with very low latencies.

Moreover, it interacts seamlessly with Java code, allowing the development of “PulseApps” (applications that run on top of Pulse), and its easy and quick integration with Java enterprise software.

A highly expressive data manipulation and querying language (PulseQL).

Based on standard SQL and with improvements from efforts like the Continuous Query Language (CQL) and LINQ, PulseQL raises continuous queries to an entirely new level. It allows easily define temporal queries, continuous values, event-based dictionaries, joins among different types of entities and windows (time or tuple-based), in an intuitive and expressive way.

Off-the-shelf definition of Metrics and Key-Performance-Indicators (KPI)

Visualizing information in real-time is good, but data needs to be persisted and queried. Moreover, in order to be productive the level of abstraction must be raised beyond low-level queries.

Pulse allows to easily define metrics that are both available for real-time visualization and data treatment, but are also aggregated, consolidated and persisted. Once a metric is given a name it can be uniquely and transparently referenced from all Pulse modules.

Predictive Analytics though a state-of-the-art Baseline Generator

The Pulse baseline generator allows you to instantly know if the metrics currently being monitored are within expected ranges or not, and how they are expected to evolve in the future.

FeedZai Pulse automatically saves aggregated information regarding key-performance indicators and, given a baseline profile, automatically calculates the current expected values as well as the baseline curve for the upcoming hours. Multiple baselines can exist for the same metric and support for metric and baseline hierarchies are native to the product.

Alarms and Actuation

By combining key-performance indicators with a flexible definition of baselines, alarms and actuation can be automatically raised when something unexpected happens even when no human is monitoring the system. FeedZai Pulse allows to continuously monitor, up-to-the-second, hundreds of key-performance indicators and baselines, over large data hierarchies, acting when it matters the most.

Middleware Integration

FeedZai Pulse was designed to integrate easily and efficiently with existing core IT systems. It provides adapters for wide variety of systems, including:

  • Java Messaging Service (JMS), allowing direct integration with Enterprise Service Buses (ESB) and Service Oriented Architectures. In particular, it also supports direct integration with IBM Message Queues.
  • Database Systems through JDBC, allowing collecting and send information into existing operational databases and also data warehouses.
  • TCP/IP Sockets (Binary and CSV), allowing sending and retrieving information directly from a communication interface.
  • File Interface (CSV), supporting reading and writing directly from CSV files with very low overhead.
  • HTTP/JSON, proving support for passing directly through firewalls and using a REST-type of approach for data integration into the enterprise.
  • XML-RPC, which provides a foundation for creating fast web services in an easy and configurable way.
  • Java Management Extensions (JMX), making the core information regarding the several modules of Pulse readily available for the IT management infrastructure.

Furthermore, Pulse provides a well-defined input/output API which allows to easily connect to other systems for which adapters are not available, or to create Web Services that expose the information being produced


Data Visualization

FeedZai Pulse is used in different ways. While many customers only require its advanced data processing functionalities, which are used close to their core IT systems, others want to be able to see results in real-time on astonishing dashboards.

Pulse Screen

The Pulse provides a visualization layer:

  • Fully compliant with HTML5 and JavaScript, supporting all major browsers in the market (IE, Firefox, Chrome).
  • Provides a pallet of Rich Graph types (line graphs, bar charts, closed areas, among others), all updated in real-time.
  • Supports Dial-charts and Bullet-Graphs updated in real-time and integrating both historical data coming from baselines as well as real-time data coming from streams.
  • Live-Drill over real-time data and Real-Time OLAP. Users can easily drill-down and drill-up over streaming real-time data, combining visualization on a tabular format with corresponding will-down in real-time streaming graphs.
  • Fully integrated with JSP/Servlet/JSF Development. FeedZai Pulse ships with an embedded fast web server (Jetty), which support all the later specifications regarding Java web development.
  • Pulse web-enabled applications is greatly simplified.
  • Very flexible deployment. Since the core visualization components of Pulse are based on HTML, the visualization layer can be deployed in all major web servers (e.g., Apache, Microsoft Internet Information Server, Jetty, Tomcat).