“Big data” is undoubtedly one of the biggest technology buzzwords right now. Hype aside, its application has huge transformational potential for businesses and even society as a whole. The impact of big data is being felt in industries ranging from healthcare to e-commerce, as well as in non-commercial sectors, including political campaigning and law enforcement. Organizations that aren’t yet in on the act are busily trying to decide if they should be, and, if so, what exactly they need to be doing. Ultimately, though, accumulating big data isn’t the point. What are important are the big answers the data can provide, or more precisely, the application of big data technology and analysis techniques to make smart decisions.
However without the proper framework to store, curate, visualize, and understand vast data sets, data is just so many ones and zeros clogging up countless petabytes of hard disk space the world over. With the sheer volume of data involved, traditional data-handling techniques are rapidly becoming obsolete, and a new economy of big data has emerged with the technology to match. It’s in this context that Intel recently announced its own entry into the world of enterprise big data, with the Intel® Distribution for Apache Hadoop* software solution (Intel® Distribution).
Based on the Apache Hadoop open-source framework, the Intel® Distribution offers a way to store, manipulate, and analyze vast volumes of data through a distributed cluster solution. The Intel Distribution has been optimized for Intel® server hardware and delivers major efficiencies, all with the goal of helping organizations arrive at their actionable destination more easily and quickly.
Intel is using its world-leading hardware expertise to deliver a solution that can dramatically reduce processing time. In one benchmark test on a system based on Intel® Xeon® processor-based servers, solid-state drives, and fast network adapters, the Intel Distribution reduced the time needed to sort and analyze one terabyte of data from an average of four hours to just seven minutes.1 To any organization working with big data, achieving that level of time saving is a clear benefit.
The four main features of the Intel Distribution together create a compelling big data enterprise solution.
- Security. Intel delivers an unprecedented level of encryption and decryption at the processor level on its Intel® Xeon® processor-based hardware with little noticeable slowdown2.
- Performance. The Intel® Active Tuner for Apache Hadoop software automatically optimizes cluster performance on-the-fly, resulting in the Intel Distribution delivering a performance gain of up to 40 percent compared to other solutions3.
- Intel® Manager for Apache Hadoop software. This software provides all the tools users need to get the answers necessary for optimum decision making.
Big data is here to stay, and its impact is being increasingly felt in business and beyond. With the Intel Distribution, organizations can master the use of big data, unlock big competitive advantages, and ultimately unleash its full potential for positive change.
For More Information:
Check out the Intel Distribution introductory video and additional Big Data related videos here.
Watch a video of the Intel Big Data Launch keynote from February 26th, 2013, here.
For everything to do with Intel and Big Data, visit the Intel Big Data portal here.
1Based on internal Intel tests
2AES-NI whitepaper: https://hadoop.intel.com/pdfs/IntelEncryptionforHadoopSolutionBrief.pdf
3Presentation by Intel research scientist Kushal Datta at Strata 2: http://www.youtube.com/watch?v=PBj5gMxGYRg
