Good times with fast data
What is fast data? A few familiar examples include clickstream, sensor data and video. With real-time analytics, this data will not only move fast, but provide fast insight to ensure your organization will always be celebrating.
We live in an era of now! Twitter, Pinterest, videos on demand, mobile devices, pop up ads, texts, smart meters, wearable healthcare devices and much more can either make our lives better, with appropriate context to suit our needs and interests, or a nightmare of chaos and noise.
The challenge for big data technologies is to make sure these fast data types turn into high performing sources of competitive advantage. With the release of the IBM Institute for Business Value Report, Analytics the Speed Advantage, we know this is top of mind for our clients and business partners.
Ovum first coined the term fast data, a few years ago back in 2012. Fast data doesn’t refer to a particular type of data or technology. Instead, it’s data moving at very high speeds. What’s innovative is new big data technologies that can capture this data in motion and analyze it to address top business challenges like:
- Internet of Things (IoT): Optimize availability, performance, capacity and resource utilization
- Enhanced security intelligence: Predict, prevent and act on security threats and real-time fraud detection; increase situational awareness
- Next best action: Act on up-to-the-second observations, while the event or transaction is still happening
- Real-time sentiment analysis of social media: Effectively respond to improve the client experience
There are different kinds of technology available to analyze fast data including in memory databases, complex event processing or operational intelligence solutions. But there is only one technology (stream computing) that achieves the best performance and is the broadest adopting according to research from Evans Data Corp. In fact, 18 percent of surveyed developers are using InfoSphere Streams to harness fast data to turn it into meaningful and actionable insights.
IBM InfoSphere Streams takes this fast data and delivers great times in terms of performance, scalability and simplicity. It does analytics faster, on a smaller hardware footprint (up to 14.2 times less hardware resources and up to 12.3 times more throughput) with a development platform that enables 45 percent faster delivery of real-time applications.
Instantaneous responses are required for stock trading, national security or for disease detection. But it is important to realize that a fast response without power analytics to back it up is not as valuable. A quick response time on a query without the deep analytics (including geospatial, text and entity analytics) has a more limited application.
The way to address the challenge of fast data is to continuously perform analytics on data streams all the time. Use statistical models on data in motion that are constantly changing to respond immediately. This compliments existing data at rest analytics solutions.
So, do you have good times with your fast data?