Infotech Lead India: The hottest trend in data warehousing is the growth of big data, which refers to the rising volume, velocity and variety of digital information – in every business. The rise of data volume worldwide is predicted to increase to 35 trillion gigabytes by the year 2020 – much of it coming from new sources including web logs, social media, mobile phones, and sensor networks. Organizations are racing to build data management platforms capable of managing the huge riptides of data that stream into their businesses every day. The future of many businesses will be determined by how well it captures, analyzes and turns this explosive growth of data into revenue and competitive advantage. , talks about trends in the industry. Excerpts of an online interview:
What benefits do you bring to organizations in terms of ROI and performance?
Organizations that use Teradata are reporting triple digit returns on their Teradata investment. Specifically, the State of Michigan is reporting one million dollars a day ROI on Teradata. This is a well-documented case from Michigan’s Director of Data Center Operations, Carol Sherman, who first reported the ROI in an open session at a Teradata PARTNERS User Group conference in October. The ROI figure of one million dollars a day, translates to $41,667 dollars an hour or about $700 a minute. That’s how much and how fast the state of Michigan has benefited by integrating and centralizing much of its data in a Teradata integrated data warehouse (IDW).
For commercial organizations, the ability to transform data into competitive advantage is a critical success factor today. Early in Teradata’s history the only way we could sell was if we proved the ROI first. That stuck in our DNA, it’s a part of our culture. If you read Davenport’s book “Competing on Analytics” you will see Teradata customers prominently described with powerful ROI results. Harrah’s, Bank of America, Bouygues Telecom, Capital One, GE, Dell, WalMart – the successes they have had using Teradata are astounding.
What are the most attractive industry verticals for you?
To be frank, the attraction is actually not in any one industry. Teradata serves more than a dozen industries – and it would be off-point if I tried to tell you that one industry is more attractive than another. Each industry has an appeal. For example, Healthcare is growing the fastest as is the Utilities industry. These are vibrant markets we are working with. But there is no way we could de-emphasize banking, telecom, manufacturing, or retail – the historical big buyers of analytic databases. If anything, we are more interested than ever in mid-sized companies and emerging markets like Russia, China, and the Balkan countries.
With more data flowing from diverse networks, security concerns are high. How do you tackle the worries of your customers in this regard?
Teradata has created and patented an architecture for protecting personal information and other confidential data within a database. Teradata was a founding member of the International Security, Trust & Privacy Alliance (ISTPA) that has produced a Privacy Framework that serves as the basis for developing products and services to support privacy regulations and business policies. Also, the Teradata Database supports a robust set of fully integrated security features. For example, we recently released row-level security that allows every record in the database to be protected even though the user may have high security clearance to most of the database. There are numerous security features in both our hardware and software including user authentication, LDAP integration, discretionary access control, mandatory access controls, role-based access controls, strong data encryption, and disk level encryption.
Dealing with unstructured data is a serious challenge for enterprises. What is your solution?
Teradata’s new Aster Discovery Platform is the best solution on the market. The Aster MapReduce Analytics Portfolio provides a suite of ready-to-use SQL-MapReduce modules to analyze unstructured and multistructured data — and accelerate data science application development. Aster has made it simple to all forms of statistical analysis – patterns, time series, market basket, graph analysis, and. Our patented N-Path function is fabulous – it enables SQL-MR to look at multiple forms of unstructured data (Twitter, Facebook, web logs) across multiple-channels in time-series and detect patterns all in one sweep through the data. It allows the data scientist to find new rules predicting customer churn, new kinds of fraud, and helps optimize digital advertising spending. What used to take days or weeks can now be done in minutes or hours. Fabulous!
Any major technology enhancements in pipeline?
The ink has barely dried on our recent releases of our hybrid columnar technology and Teradata Virtual Storage. Our hybrid columnar ‘leap-frogged’ numerous competitors to provide both row-storage and column-stores in the same relational table. This means the customer can design their database for faster performance and high data compression without any shackles. You’ll notice that the hype around who has the best columnar storage died down about the time we released our solution.
We are also delivering 100 percent automatic data migration from hard disks to solid state drives (SSD) based on the data temperature. Teradata Virtual Storage is highly sensitive to the business user demands. The more the business users ask for specific data, the more we migrate it to ultra-fast SSDs. We can do this because our database software knows what kind of data it is, who has been using it, and how often a query needs that data. Other solutions operate only at the hardware level with no knowledge of what the data is or how it is really used. It’s a great competitive advantage.
Looking into the near future, watch carefully for our next release of mixed workload management and enhancements to our multi-temperature data strategy. These innovations will delight our customers and frustrate our competitors.
Can we expect any major partnerships in the near future?
We partner with hundreds of vendors; one of our top partnerships is with SAS, the analytics giant, because analytics is extremely important to Teradata, and most of our customers use SAS in some way. But we also partner strongly with IBM, Oracle, and SAP – it’s the classic ‘co-petitio’n arrangement where some products complement each other while some compete.
We have recently made significant R&D investments partnering with HortonWorks on big data Hadoop technologies. Our SQL-H capabilities in Teradata Aster are a direct outgrowth of that partnership.