Many organizations today realize the importance of big data in the business world, but although many have gotten better in processing and storing the range of data rushing in from a myriad of sources, problems nonetheless persist.
One of the major issues for companies trying to leverage big data is the length of time it takes for data to be analyzed. While being able to gather and store the data is essential, big data is useless if it cannot be analyzed. As data continues to grow, the processes for moving and analyzing it only become slower and more tedious.
To look at the new tools and technologies available DBTA hosted a webcast featuring Mark Theissen, CEO of Cirro; Peter Hoopes, VP/GM of Birt Analytics Division Actuate; and Amit Patel, program director, Data Warehouse Solutions Marketing of IBM.
There are three trends concerning faster analytics: the modernization of enterprise architecture for analytics, data lakes, and on-demand distributed analytics, said Cirro’s Theissen. “Most enterprise infrastructures and architectures are really ill-equipped to deal with the challenges of big data,” he explained.
With the many new types of data and emerging processing platforms, it is difficult for business analysts to process data efficiently. One approach - the data lake - varies from vendor to vendor depending on each organization’s purpose for their own data lake. Data lakes initially were developed as a means for cheap scalable storage for companies. As time has gone on, data lakes have emerged as a storage entity where organizations can place their data and perform analytics. While the concept of a data lake is good, in practice, enterprise data lakes often are complex and fragmented. On-demand distributed analytics is the concept of analyzing data in real time. This becomes difficult with multiple real-time data sources. Cirro, Theissen said, provides the option of federating the processing across the different data sources for its customers.
Actuate’s Hoopes discussed how analyzing big data in real time is an issue for business analysts. Hoopes cited a recent Unisphere Research study that found that “50% of organizations today are spending too much time finding the data they need to analyze.” Speed matters when it come to sifting through data and being able to process data quickly is key in being able to analyze it quicker than competitors. To quicken the process, Actuate believes in iteration. This allows getting answers quicker and then asking new questions to get deeper into the data.
In-memory computing can help, added IBM’s Patel. A major difference between in-memory computing and traditional methods is that the data is stored in RAM. IBM’s new software BLU Acceleration contains four elements that help in speeding up business analytics: next-generation in-memory, analysis of compressed data, CPU acceleration, and data skipping. BLU Acceleration also offers BLU Shadow Tables which increase the speed of analytics. “All of this results in three benefits for the customer; fast, simple, and agile,” stated Patel.
A replay of this webcast is available from the DBTA website.