• Data Your Way.

    Data Your Way.

    It's not the size of the data, it's how you use it.   Big Data technologies may be great at collecting large amounts of diverse information but to ask any meaningful questions of the data, analysts will first need to get it into a format their tools can understand. As enterprises shift focus from capturing and managing data to actively using it, providing high quality context-driven information remains a challenge. Solving this problem requires a new set of tools that specialize in agile data preparation, allowing users to quickly assemble made-to-order datasets.
  • 1

SERVING RAW DATA WITHOUT ETL

ANALYTICS 3.0

Remarkably, almost 80% of the time dedicated to data analysis is actually spent on data preparation. Less than 10% is spent modeling and looking for patterns. While data is ingested into analytical apps faster from an increasing number of diverse sources, data quality and freshness are often lagging. Noisy, unreliable information affects business decisions and has a direct impact on the bottom line.

Although a number of new big-data solutions have recently emerged, the goal of these technologies is primarily storage of high velocity data. For many enterprises, their ability to collect information has surpassed the ability to quickly mobilize data for analysis. Resulting architectures are often a fail-back to the traditional warehouse model with Hadoop or NoSQL thrown into the mix.

At StreamScape we believe it is the lack of agile data preparation ability that leads to so much back-pedaling and frustration. After all, new decision support capabilities can’t be developed using old, rigid analytics models.

Traditional data analysis processes extract data from a raw source, transform it into target schema possibly requiring flattening or condensing, and then load it into a database or other data management system. This approach is unsuitable for timely analysis. Lag time introduced by ETL makes for stale data. More importantly, data management systems organize information for optimal storage using a single, well-defined schema. Changing schema to benefit one application results in change to all apps, making maintenance and deployment a slow, complex and expensive process.

Alternativley the StreamScape query engine can work with raw data and apply schema on-the-fly based on application needs. This eliminates the up-front cost of data ingestion, allowing users to describe, prepare and present data to consumers as it is being created - straight from the source.

The Reactive Data Platform™ functions like a data factory, separating preparation tasks into data modeling and an assembly pipe-line. Rich semantics and ontology let users define, query and document relationships between disparate information sources, making it possible to create many views of the same data as needed, without impacting other applications.

Data analytics have evolved from simple, fact-gathering and reporting tools into application platforms that scrub and package information, turning it into high-value product. Analytics encompass a wide array of software used to collect, analyze, and disseminate data for purposes of better decision making.

Using analytics for competitive advantage is not just for Google, Amazon or Facebook anymore. Innovation in data technology, combined with low-cost cloud infrastructure has made it possible for any company to capitalize on vast new sources of information. If your business makes, moves or consumes things, works with customers or partners, you have increasing amounts of data that can be analyzed to get an edge on the competition, or create new products and services. That's the essence of Analytics 3.0.

Competing on analytics means competing on technology. So data-smart companies invest in solutions that let them make the best use of available information, wherever it is. Performance is not always the goal. Sometimes it makes sense to move analytic computations to the data and other times the opposite is true. SQL queries may work best on simple files. Complex JSON or XML content may be better off converted to streams and analyzed that way. Without proper technology to make data nimble and adaptive, analysis often becomes a complex and expensive coding exercise.

Adaptive data systems are at the heart of next-generation analytics. Building data intelligence into the products and services that customers buy (RFID or similar sensors) is a start. Exploiting the information in real-time by aggregating streams of sensor data and joining them to historical records across multiple back-end systems is another feat. Companies that want to prosper in the new data economy must fundamentally rethink how analysis can create value for them and their customers.

Supporting tools must handle all such requirements, should easily complement a firm's existing infrastructure, have a light footprint, be scale-able and cost effective. A set of capabilities the StreamScape platform successfully delivers on.


[No form id or name provided!]
.. get started with
the Reactive Data Platform™ today!

  Login

Login to access additional content such as white papers, on-line docs, Wiki and product downloads.