At Strata Conference last month, my Wikibon colleagues and I had the pleasure of chatting with Anjul Bhambri, Vice President of Big Data in IBM’s software group. While we had a pleasant discussion, I expressed to Anjul that we as an industry are waiting for IBM to show a higher level of leadership in the Big Data space more befitting a company of Big Blue’s stature and expertise.

My point was that IBM has a very broad and fairly deep portfolio of Big Data-related products and services – including its analytics services team, Watson unstructured analytics system, Netezza data warehouse appliance line, Hadoop-based Big Insights platform, and DB2 InfoSphere Warehouse – but that the company has not done a very good job of explaining to its customers or the market how all those pieces fit together.

Unfortunately, that trend continued today with IBM’s announcement of DB2 10 and InfoSphere Warehouse 10.

To give IBM its due, there are a number of important feature updates in this release. They include:

• Improved adaptive data compression capabilities, including table-level and page-level compression, resulting in lower storage requirements and faster query performance in I/O bound environments.
• Automated tiered storage based on data “temperature,” meaning hot data, or data in high-demand, is stored on SSD RAID for frequent access, warm data is directed to SAS RAID for occasional use, and cold data is stored on SATA RAID for infrequent use.
• Support for graph store functionality inside DB2, including support for SPARQL 1.0 query language for RDF, which IBM says accelerates rational use case by up to 3.5 times.

Another improved functionality IBM touted during its call with analysts this afternoon was the ability to “trickle” feed data from transactional systems into the InfoSphere platform to allow for near real-time data warehousing. This got me thinking about Netezza, and how IBM differentiates the two platforms. I asked for clarification during the Q&A period.

The first difference to understand between Netezza and DB2, I was told, is form factor. Netezza is an appliance-only product and allows for very little if any customization. DB2, as a software-only DBMS, is highly configurable. The second point of differentiation is in the types of workloads each is designed for. Netezza is designed for highly analytic workloads, while DB2 is able to handle more read/write intensive, transactional workloads, in addition to analytics.

Both are perfectly fine answers, but we knew them already. I was expecting – and I know customers and members of the Wikibon community are expecting – a crisper, clearer response from IBM when it comes to explaining how the many pieces of its Big Data puzzle fit together. What are some clear use case examples that illustrate when one data warehousing platform is more appropriate than the other? When is Big Insights or InfoSphere Streams a better fit? How do all these platforms communicate and interact with one another? And how does IBM’s services team help customers make these distinctions and otherwise navigate the Big Data landscape?

Admittedly the job of rationalizing and communicating its approach is harder for IBM, with so many Big Data tools and services in its portfolio, than a Big Data pure-play with just one product. But enterprise and SMB CIOs alike are practically crying out for leadership in the Big Data space. Specifically, in addition to the above questions, CIO’s want to know (and should be asking IBM at every opportunity):

• Why should I invest yet more money and resources in technology to gain insight from data when I’ve already got a large, complex (and expensive) data warehouse platform (be it form IBM, Oracle, or Teradata)? In other words, IBM, explain the Big Data Big Picture.
• How is Big Data applicable to my vertical market and, specifically, what services can IBM offer me to help identify the Big Data use cases, both vertical and horizontal, that will return the greatest business value?
• How will Big Data technologies from IBM, like Big Insights and Netezza, fit into my existing information management infrastructure and how can IBM help me ease the transition with minimal disruption?
• With a well-known shortage of Big Data talent available, how can IBM services help me build out my Big Data practice from both a Big Data infrastructure and Data Science talent perspective?

IBM probably has the most comprehensive portfolio of Big Data products and services on the market, but it needs to do a better job of telling a complete story. Today’s DB2 10 announcement embodies IBM’s Big Data communication struggles: The company did a great job highlighting specific functional improvements, but failed to put them in context or explain how they fit into the Big Data Big Picture.

Leaders need to lead.