Interview with Daniel Graham, General Manager, Teradata
Data is probably the biggest asset for an organization in today’s digital economy. But, with huge data, comes a challenge to mine it quickly and effectively. To solve this problem, Teradata provides data warehousing and analytics solutions across various sectors to help companies make better, faster and smarter decisions. Having launched its IPO in 1987, Teradata is one of the largest companies in providing best in class big data analytics and marketing applications.
I personally experienced Teradata solution some time back, when a move to Teradata (from one of the competing solutions) reduced the query execution time to a fraction of previous time. I have ever since been impressed by Teradata’s offering and performance.
Recently, we got a chance to interact with Daniel Graham, General Manager – Enterprise Systems, Teradata who has been working in the this company for more than a decade. Dan is responsible for strategy, go-to-market success, and competitive differentiation for the Active Data Warehouse platform and Extreme Performance Appliance at Teradata.
Below is the transcript of the our conversation:
AV(Analytics Vidhya): Thanks Daniel for devoting time to speak with us. It has been 9 years since you have re-joined Teradata. How has this journey been for you? How has this company transformed from your early days with them in the 1990s?
DG(Daniel Graham): The big turning point for Teradata was the spin off from NCR Corporation in 2007. The spinoff allowed Teradata to expand into multiple areas. The Teradata portfolio is much broader now, yet still focused on analytics. Our transition from a single product line to a broad portfolio of products and services has been dramatic.
For example, today we have appliances at every price point targeted at specific workloads such as discovery or regulatory compliance. The spin off also freed up funding for innovative software investments such as Active Data Warehousing and Teradata Virtual Storage. But most importantly, the spin off allowed an independent Teradata to acquire other companies.
Our acquisitions –AsterData, Claraview, Aprimo, ThinkBig, Revelytix, and others–are strategically known as a string-of-pearls. We’ve had great success retaining the people and vision of the companies we acquire. Over this same period of time, Teradata Professional Services has expanded exponentially to over 5,000 data quality, performance, security, data mining experts.
AV: Give us a brief overview of products offering by Teradata?
DG: Our hardware portfolio consists of four Teradata Database appliances, one Aster Database appliance, a Hadoop appliance, a SAS appliance,and a Teradata Cloud offering. These are hardware platforms that are configured to match to specific customer workloads.
Our software portfolio begins with the Teradata and Aster Database engines. To this we added the Aprimo Marketing Resource Management, Omni-Channel Marketing, and Digital Marketing solutions.For Hadoop systems, we offer Teradata Loom (a file management and data wrangling solution) and Teradata Rainstor (a big data archival system).We also offer Teradata Analytics for SAP.
AV: That is quite extensive for any datawarehousing need. What are your focus areas currently? What can we expect from Teradata in coming days?
DG: Today, we are rolling out systems that provide maximum scalability, high performance analytics, and support hundreds of concurrent users. Specifically, the Teradata’s Data Warehouse Appliance 2800 is the most robust database for analytic data warehousing. This appliance’s innovative design is optimized for fast scans and deep-dive analytical workloads. The Teradata Enterprise Data Warehouse 6800 gives leaders in every industry a fresh, big-picture perspective. With a 360-degree view of the business, the organization can gain valuable insights into customers, supply chain, and financial and performance management – from the boardroom to the call center to the loading dock.
Teradata is also putting a lot of resources into Teradata QueryGrid to enable massively parallel queries to run across Hortonworks, Cloudera, Aster Database, MapR, Oracle, and MongoDB platforms, among others.This allows a business user to submit a SQL query on one platform and join data together with other platforms – all at high speeds with no extra programming required.
We are also expanding Think Big – a Teradata Company – into Europe. Think Big is our Hadoop and NoSQL professional services arm.In this same vein, we also announced Think Big’s Dashboard Engine, the industry’s first unstructured data OLAP solution based on Hadoop.
AV: How do you ensure that your customers stay on latest offering from Teradata?
DG: There is a hardware and a software answer to that question. For software, the best enticement for customers to stay current is new features that they really want, which is guided by customers on our Product Advisory Committee. In addition, every new software release, yields a performance improvement, which is always welcome. For hardware, the benefits are dramatically visible. New Intel microprocessors, larger memories, and faster devices like Infiniband and SSD disks are incredible when it comes to performance boosts. Successful data warehouses always have more users and more demand for more queries, so the need for speed is eternal. Consequently, many of our customers combine a hardware upgrade with a database software release upgrade.
AV: Since, privacy and security of data has remained a major concern for companies, what innovations/modifications do you plan to bring in data warehousing?
DG: Privacy is a policy that our customers must define for themselves. Our Professional Services consultants can help customers define policies and implement some of the decisions. But ultimately, ethical behaviour – aka privacy — is a customer business decision which is often guided by country-level laws.
Technologically, Teradata has invested heavily in security over the years with some cool recent developments. Along with all the extensive authentication and permission controls in the Teradata Database, we’ve implemented row level security and secure zones. Row level security allows for the tagging of individual records, such that only people authorized can see them. For example, we could grant access to the SALES_TRANSACTIONS_DATA to everyone, but each person could only see details from their own region. Sales managers could review deals in their own region, but not neighbouring regions. You don’t have to use a lot of SQL Views to do this, just in-database row level security.
Second, we recently added secure zones. Secure zones make it appear to the user as if they were in their own isolated database on a separate machine. Imagine a multinational e-commerce retailer. Because of country laws, data from one country cannot move between countries, protecting consumer privacy.With secure zones, we can use the same database design for every country and ensure that no one outside of a country can see any of the data – not even the table names.Secure zones are also a prerequisite to multi-tenancy cloud computing.
And I can’t forget to mention our partnerships with several best-of-breed security vendors that either encrypt or mask the data in the database – IBM Guardium, Voltage, HP/Vormetrics and Protegrity.For example, our partnership with Protegrity adds encryption in the applications processing layer so that personally identifiable information cannot be seen – not even by the database administrators.
AV: Do you plan to capture the untapped market of front end solutions i.e. Data Visualization?
DG: Teradata follows a best-of-breed strategy, which means we rely on partners who represent the best in their domain. For example, we partner with MicroStrategy, Qlik, SAS,Tableau, Tibco Spotfire, as well as the large stack players for visualization and BI tools. There are times we overlap a little with these partners. But we have no intention of becoming a BI tools or visualization supplier. Our customers prefer to use best-in-class tools with Teradata just as they have decided that Teradata is best-in-class for data warehouses.We will stick to our core competencies.
AV: Qlik, in addition to an amazing data visualization software, also has a data warehousing solution(although less effective). How do you plan to compete against them?
DG: Teradata does not compete with Qlik. Teradata sells massively parallel database machines and marketing applications. Qlik sells business intelligence visualization and dashboards. In fact, we have recently co-developed new in-database processing with Qlik. Branded ‘Direct Discovery’ from Qlik, they can configure what data is placed in memory (e.g., small reference tables, metadata), and then what sources the user will drill through to the database to speed up performance (e.g., large fact tables). We think our joint customers will appreciate this hybrid approach.
AV: Follow through questions on market competition, how do you plan to compete / complement against NoSQL data solutions?
DG: The majority of NoSQL products are operational databases. These products are good for building online applications whether they are e-commerce, call center apps, or social media applications. These are good sources of data for the data warehouse. For example, a NoSQL website collects buyer clickstreams in response to online offers. It’s the data warehouse’s job to combine that data with past history, customer preferences, payment history, and other demographics. The analysis is passed back to the NoSQL application. The NoSQL application then knows to inform the business user to upsell the consumer and to minimize sales to those that default on payments. Teradata does not compete with NoSQL vendors; we embrace them as partners inside the IT architecture.
AV: What is the best part about the culture in your company? How has it evolved over the years?
DG: Teradata’s culture began in the 1980’s in California. Every one of us believes deeply in the vision of leveraging data warehousing, analytics and big data to help our customers better run their businesses. Thirty years of BI/DW market growth reinforces our sense of purpose, vision, and just good fun helping customers. We strive to be our customers’trusted advisor. And that bubbles over into mutual trust and respect. We are that team rowing in the same direction with our customers. This isn’t a platitude crafted by HR. It’s the way we are inside Teradata. It’s a fun place to work.
AV: What would be your advice to aspiring young folks seeking to build their career in data analytics/big data analytics and other related domains?
DG: Analytics is a force of nature in the digital world. And, we can expect 30 additional years of amazing technology advances. Those individuals with skills and expertise in a vertical industry combined with analytics will be the most attractive to employers. For most people this means proficiency with tools such as Tableau, Qlik, MicroStrategy, SAS and others. For some, a deep dive into predictive analytics – graduate level mathematics – can lead to a career in data science. And in both cases, skills in database and data warehouse concepts are a big plus. I also recommend reading Tom Davenport’s Competing on Analytics as well as Jill Dyche’s The New IT. These will illuminate what a career in analytics is like.
You can pick up analytic skills at Northwestern, University of Georgia, Ohio State, Stanford, and others with a data science curriculum.
AV: Thanks Daniel for that insightful and enriching interview. This interview definitely explains the vision and passion behind these awesome Teradata products and the powerful performance they enable. Once again, I thank you for your time and sharing your perspective.
P.S – We have more such interesting interviews coming up ! Do subscribe.
If you like what you just read & want to continue your analytics learning, subscribe to our emails, follow us on twitter or like our facebook page.