Expect Innovations in Big Data Analytics
Teradata, an analytic data solutions company, announced that it has identified five key big data analytic trends that will shape the business landscape in 2013 and beyond.
"The future of business belongs to those enterprises that embrace the big data analytics movement and use it to their advantage," said Scott Gnau, president, Teradata Labs. "Teradata is committed to helping customers simplify the mysteries of data and analytic technologies. Business leaders and chief information officers (CIOs) who are the quickest to adopt big analytic solutions in a unified architecture will be the most competitive. Successful companies are already extending the value of classic analytics by integrating cutting-edge big data technologies and outsmarting their competitors."
Teradata has pinpointed the five big data analytic trends for 2013:
1. The Rise of the Big Data Discovery Platform
The discovery platform will become an indispensable part of big data strategy. It provides knowledge workers — including business analysts and data scientists alike — with a reliable workbench from which to explore and perform experiments on big data, at scale, at a fraction of the time and cost required with traditional approaches. This capability has traditionally required up-front data sampling and modeling, as well as specialized skills. Discovery platforms allow companies to innovate on analytics by testing hypothesis and "failing fast" to uncover new insights in data. In addition, the discovery platform "lets the data speak;" this dialogue between the data and knowledge workers enables the business to identify new trends and insights that can lead to benefits like better consumer personalization or fraud detection.
A discovery platform must support a variety of interfaces in a single platform, including Structured Query Language (SQL), business intelligence tools (BI), statistics, and next-generation MapReduce analytics. In contrast to traditional systems, a discovery platform needs to impose very few requirements on how the data is modeled so that businesses can easily and quickly combine new and existing sources of data to speed up the discovery process.
2. Explosive Big Data Application Growth
The number of big data applications will explode over the next three years; the growth will start in 2013. Development of these applications will present challenges to CIOs because the skills required to develop big data applications are different — and more sophisticated — than the skills required to develop traditional applications. In the future, big data will be consumed by knowledge workers and applications alike. This new generation of applications, including Web and mobile, will be powered by big data insights in multiple industries. This will drive a huge competitive advantage enabling business to clearly see new opportunities to engage with their consumers.
3. From Fragmentation to a Unified Architecture
The variety of new big data technologies and platforms from which to choose will be both a blessing and a curse. In 2013, some organizations will deploy the new big data platforms in an IT environment that lacks a unified architecture and does not integrate data, metadata, security and administration. The use of big data analytic point solutions in a fragmented IT environment can kill the promise that big data can provide better insights by doing more analytics on all data. Such deployments will quickly lead to a torrent of failed big data projects.
It doesn't have to be that way. Tighter integration of technologies that support enterprise standards and leverage existing investments in analytical tools is essential for the success of big data initiatives. Deploying applications in a unified enterprise environment makes analytics simpler, faster and more powerful; while reducing deployment and operational costs. The new analytic capability provided by the applications can catapult an organization forward.
4. Blending of Capabilities
To be competitive, organizations need the capability of both big data analytics (MapReduce and procedural analytics at scale) and traditional analytics (SQL) that run within a relational database management system. As a result, big data analytics will not come close to replacing traditional analytics in 2013. The debate about which one will replace the other is unproductive. CIOs and business users will begin to blend the capabilities of the two to meet the intelligence needs of the business. Tools and technologies that provide a native blending of classic and new data analytics techniques will have an inherent advantage as the market realizes their value in 2013.
5. Storage is Not Enough
CIOs will move beyond focusing on hardware for storage of massive amounts of diverse big data to developing an analytic process that is repeatable and provides business value. Then, CIOs will be able to transition away from buying point solutions to deploying big data platforms. Advanced technology is now field-tested and available that can store and transform massive volumes of diverse data into usable intelligence for deployment across the organization. CIOs will look to their existing information management vendors to provide innovation without disruption. The CIO and their organizations that are the quickest to adopt the now-available big analytic solutions will be the most competitive.