By Charles King, Pund-IT, Inc. July 9, 2014
The latest of IBM’s annual Big Data & Analytics Analyst Summits held recently in New York City came at an interesting time for the company, the analytics community and the larger IT industry. Analytics and “big data” technologies have become central to many vendors’ go-to-market strategies. At the same time, public and private sector organizations are adopting related solutions in hopes of capturing more insight and value from their continually growing information assets.
For IBM, it’s been just over a decade since the company began expanding its core business intelligence holdings to incorporate analytics assets and capabilities. That effort cost well over $15B in external acquisitions and internal organic development resulting in key breakthroughs, including the “Watson” platform that thoroughly drubbed the all too human Jeopardy! champions in 2011.
The commercial results also appear to have been well worth the cost of those investments. In New York, IBM executives noted that the company is on course to achieve the goal it set for itself in 2010—to build analytics and big data solutions and services into a $20B annual business by 2015. But at the same time, we noticed a subtle shift in IBM’s strategic messaging around analytics and big data.
The Four “V’s” Revisited
IBM sums up its strategy in what it calls the Four V’s of Big Data:
- Volume – highlighting the continually expanding scale of business data which is growing at nearly 60% per year
- Velocity – representing both the increasing speed of data movement and the need to analyze information more rapidly
- Variety – reflecting the massive impact of managing/analyzing the unstructured and semi-structured information that makes up about 90% of the data businesses create and store
- Veracity – focusing on the critical role that data quality and accuracy play in effective analysis
Not too surprisingly, these points are similar to the strategies and offerings of other big data vendors, but the fact is that the vast majority of players focus on the first three points. Why so? Because they are relatively simple to approach and explain. Both volume and velocity generally relate to system/storage features and design that can be tweaked to enhance capacity and throughput. Variety is a code word for supporting the analysis of semi-structured items like email, reports and PDFs, or unstructured photos, video and audio files with technologies such as Hadoop and MapReduce.
Veracity, however, is another kettle of fish that is familiar territory for vendors with significant database assets (IBM, Oracle, SAP, Microsoft, etc.) and/or sophisticated information management tools and solutions (most DB vendors, plus mainline IM companies like EMC Documentum and specialty firms, including Attunity, Cloudera, Composite and Splunk).
But at the Big Data & Analytics Analyst Summit, data Veracity was front and center in numerous IBM main tent presentations, breakouts and 1:1 meetings. Why so? For a couple of reasons:
- The subject gets less attention than you might expect, especially from mainline DB vendors. This may simply be due to the fact that data accuracy and related topics are so central to core DB processes that vendors may not feel any need to emphasize them. That may be true when you’re talking with traditional DB users, but the success of big data rests on taking the message to entirely new audiences. Critics might say that IBM is simply repeating old news here (though the company did discuss interesting new capabilities under NDA), but we believe it could have a significant, positive impact on new audiences.
- It also allows IBM to focus attention on its own considerable information management portfolio, its close integration with the company’s homegrown DB2 database offerings and the roles they play in new solutions like BLU Acceleration, a powerful in-memory solution for DB2-based data.
- Emphasizing veracity features provides IBM the means to position itself as a full service big data vendor whose solutions incorporate a host of hardware, software, middleware and management offerings critical to the efforts and success of enterprise clients.
- Finally, these efforts are core to what IBM calls the “continuous curation” of data that focuses both on maintaining or improving data quality and building the contextual foundation that is so critical to realizing the value of information and building competitive advantage.
Of IBM’s competitors, only Oracle can bring as full an end-to-end solution set to bear on the enterprise market, but the company tends to focus considerably more energy and attention on traditional DB markets and use cases than it does emerging big data opportunities.
Perhaps the most radical element apparent at IBM’s Big Data & Analytics Analyst Summit was the company’s clear intention to “democratize” big data by making analytics tools and services available and affordable to nearly any business organization. That is not to say that IBM is abandoning its successful and profitable analytics solution efforts. Those will continue to inhabit and grow in enterprises across the globe.
By investing in efforts like Watson that open complex data stores and information systems to wider audiences and use cases, and by developing analytics solutions that can be delivered as services via its SoftLayer cloud infrastructure, IBM is making the power and benefits of big data and analytics available to virtually any organization. In that sense, what IBM showed analysts at its recent Big Data & Analytics Summit was nothing less than a peek through the window at a coming revolution.
© 2014 Pund-IT, Inc. All rights reserved.