With the release of webMethods Nirvana Messaging, together with its Complex Event Processing (CEP) product, webMethods Business Events and Terracotta in-memory technology, Software AG has taken a major step in implementing its management of Big Data strategy announced earlier this year. Analysis of live data streams correlated with large volumes of historical data held in-memory offers a complete management view of real time, enterprise-wide business events highlighting any divergence from normal business developments. Unexpected changes to business conditions, fraud detection in mid-transaction or dynamic risk management can result in real-time actions, pro-active operational changes or rapid process adaption. Organisations can transform their businesses, dynamically streamlining resources and addressing new opportunities.
“Increasing competitiveness and streamlining operations are vital in these uncertain economic times,” says Darren Roos, COO EMEA and APJ, Software AG. “Identifying new business opportunities, eliminating fraud, reducing risk and its associated cost is dependent on identifying what is important, or unusual, in an increasing volume of data. CEP, in-memory data management and ultra- fast messaging provides this insight to the operational business in real-time.”
Businesses face a huge increase in data volumes, which present problems of both storage and analysis in getting the right information to the right people at the right time in order to make the right decisions. At the same time, CEP is equally about maximising the value of a much broader variety of data across multiple customer interaction channels - with unstructured data such as emails, social media and mobile now accounting for as much as 80% of information held within today’s business. Speed is also critical, as the pace of business and the velocity of how fast data is being produced and must be processed has also increased dramatically.
According to the Basel Committee on Banking Supervision, hundreds of billions of Euros will need to be raised by banks to meet the new bank capital requirements. A recent report by KPMG highlighted that regulatory pressures on banks worldwide will have an impact on profitability and liquidity that will amount to a ‘systemic reduction’, if business models do not change. These changes must provide real-time transparency to all transactions, allowing the consolidated business exposure to be immediately reflected in new capital requirements. New models must also allow business misconduct to be detected and addressed as it happens.
Pattern-based event strategy
The webMethods Business Events solution achieves this by supporting a well-defined pattern-based event strategy, defined by Gartner as “offering a framework to proactively seek patterns from traditional and non-traditional sources, model their impact on business operations or strategy and adapt the organisation.”
Business patterns are “sets of recurring and/or related business activities or events” and a pattern-based strategy brings together the disciplines needed for business leaders to exploit opportunities or identify threats and respond in a timely and intelligent way. “The business case for any IT investment is typically based on either the need to make money, save money or manage risk,” says Roos.
“With our pattern-based approach, you can directly address all three key investment criteria. It does this by meeting the challenge of correlating and interpreting data held in separate, isolated silos.”
The CEP market has been growing rapidly (31% CAGR between 2008-2013), reaching around $580million by 2013. Gartner predicts that, by 2013, enterprises using system-based pattern detection in strategic planning and tactical decisions will outperform their peers by 20%, with software spending in pattern seeking and modelling likely to break the $2billion barrier this year.
Scalable, integrated approach
At the core of webMethods Business Events is a proven, scalable and highly-innovative CEP engine which is optimised for high-throughput and low-latency event processing to deliver predictable, traceable, repeatable and correct results, every time. The engine can continually monitor thousands of event streams and tens of thousands of events per second, looking for complex correlations, patterns and sequences and taking the most appropriate action when it finds them.