Data Analytics at the 'Speed of Thought'
By Susan Nunziata Pondering the announcements made so far at Oracle OpenWorld, which kicked off in San Francisco on Oct. 2, 2011, with CEO Larry Ellison's keynote address, one can't help but marvel at an oft-repeated phrase in the company's product plugs: "Speed of thought."
We heard it multiple times during this morning's in-depth presentation on the Oracle Exalytics Business Intelligence Machine, which is touted as "the industry's first in-memory hardware and software system engineered to run analytics faster than ever, provide real-time speed-of-thought visual analysis, and enable new types of analytic applications."
What, exactly, is speed-of-thought? Would that be "speed-of-thought" before or after you've had your morning coffee? Has Oracle discovered that thought travels at breakneck speed -- much like those little neutrinos which some physicists suspect are able to travel faster than the speed of light?
Sure, the technology array is impressive. According to the company, Oracle Exalytics is comprised of:
- Oracle's Sun Fire server featuring 1 Terabyte of RAM and the Intel Xeon E7-4800 processor with a total of 40 cores;
- Oracle BI Foundation software including Oracle Business Intelligence Enterprise Edition and Oracle Essbase featuring performance optimizations and an improved visualization environment for interactive analysis without limits on underlying data and classes of users; and,
- Oracle TimesTen In-Memory Database for Exalytics, based on Oracle's proven in-memory RDBMS with lightening fast response times, enhanced and optimized for business intelligence.
Even better, "it's an an open solution for use in heterogeneous IT environments, designed to access and analyze data from any Oracle or non-Oracle relational, OLAP or unstructured data sources such as IBM DB2, Microsoft SQL Server, Netezza, SAP Business Information Warehouse, and Teradata, amongst others."
For organizations seeking even faster analytic performance, the company says, its Oracle Exalytics features an InfiniBand connector designed to work with the Oracle Exadata Database Machine.
Existing reports and dashboards built with Oracle BI Enterprise Edition and Oracle BI Applications run on Oracle Exalytics without any changes, according to the company's product announcement.
The product release goes on to say the solution will enable:
- Speed-of-thought interactive visualization that enables casual users to perform free form exploration of large, dense data sets to easily spot patterns, trends, and outliers; and
- An immersive mobile experience for Apple iPad users that can scale to many thousands of mobile users, making analysis available to everyone, anywhere.
Let's think about this for a minute. What do you think will happen when casual users perform free-form exploration of large, dense data sets? Can data analysis effectively be made available to everyone, anywhere?
Don't get me wrong--I'm not arguing against the idea that we should have the ability to access the data sets we need when and where we need them. Or, that there is a desperate need for tools to help us make sense of the growing volumes of "big data" that face us in our daily work lives.
But there is a very specific set of skills required to truly make sense of massive volumes of data and -- let's be honest -- not everybody in the workplace has the, er, "speed of thought" necessary to make the most of this amount of information.
EMC CEO Joe Tucci touched on this challenge in his Oct. 3 OpenWorld keynote, during which he railed about the failures of IT in recent years. EMC CMO Jeremy Burton elaborated on those observations later today, telling us that "the amount of money [and resources] that should be spent on big data can't be spent until IT addresses its infrastructure issues."
While EMC has a clear motive for wanting to encourage migration of IT infrastructure to the cloud, that shouldn't overshadow the big-picture point here. There's no argument that many of the traditional roles and skills required in IT are rapidly changing. In the process, there's going to emerge a growing need for what Burton calls "data scientists" who are able to make sense of masses of information and then help the average business person visualize it in a way that enables them to put it to good, profitable use.