By Krishna Kallakuri - March 16, 2016
In this information age, the pace and constant changes in operating a business has never been greater. The most effective way to manage these situations is to gain “timely” insights from data that are immediately “actionable” to businesses, and where it is making a competitive difference.
Great message isn’t it? But what are “actionable” insights? Let us get to the bottom of it.
Traditional business intelligence systems do not have the ability to analyze current business performance because we already know the drill – Staging, Extraction, Transformation, Nightly loads etc of data. With this standard process the analysis is delivered to the business that is at least 1-2 days behind and that too fingers crossed hoping the ETL load are accurate and don’t fail. Between all this juggling, how would businesses react to Product life cycles that are quickly vanishing; Can they manage inventory turnover which needs to be fluid in this competitive market? We are certain that you know Amazon can now deliver products the same day. Now imagine the science applied to manage their inventory and supply chain. In this scenario, do you see a hint of actionable insight? Let us dig into more scenarios.
Fraud is a major business problem for Banking & Insurance domains. Can we solve this problem saving millions of $ in damages using real time architecture? The answer to this is a big yes. Seriously this is a no brainer. We now have scalable architectures that analyze transactional or claims data in real time and predict fraud proactively.
In retail, Dynamic Price changes presents with a business challenge to stay ahead of competition, because these days the consumers have several buying options. Ideally, the retailers would want better customer retention reacting to dynamic price changes on a timely basis. So, can we enable retailers with “timely actionable insights”? Again the answer to this is a Yes. The modern real time architectures can process enormous amounts of data whether is structured or unstructured in real time and produce actionable insights.
Network and Data Security is now a major threat for all enterprises. Is there a way to predict network intrusion before it happens? Absolutely, Real- Time architectures can take continuous streams of all network logs and run predictive models in real time to predict actionable insights. Sky is the limit to reap the benefits and define use cases for Real Time Analytics.
So what is out there that can help enterprises build these types of applications?
Storm: Apache Storm is an open source distributed real-time stream processing engine. Storm enables processing of event-based streaming data with ease-of-use using in resilient, distributed way and supports a myriad of programming language of choice.
Spark Streaming: It is a library written in Scala on top of Spark’s core API that enables spark to process streaming data in real-time. Being a library on top of Spark Core it is tightly coupled with Spark’s other libraries including Spark SQL, MLlib etc. This is an extension of Spark core and it inherits all of Spark’s features like scalability, fault-tolerance, and high throughput. Spark streaming can also re-use any of the code from Spark’s batch applications for live data processing purposes in real-time. Spark streaming provides data ingestion mechanisms from all the popular sources like Kafka, Twitter, ZeroMQ, Kinesis, TCP sockets and also from log files.
Last but not least IBM has announced its full support to Apache Spark and has integrated with their mainframe z/OS (transactional system). This is really a game changer for all enterprises running IBM zOS because they can full advantage to perform real–time analytics whether they want to prevent fraud, AML or provide next best action to customers. And all of this can now happen in real-time producing “Timely Actionable Insights”.