|By John Mertic||
|October 28, 2016 09:45 AM EDT||
Making Apache Hadoop Less Retro: Bringing Standards to Big Data
Ten short years ago, Apache Hadoop was just a small project deployed on a few machines at Yahoo and within a few years, it had truly become the backbone of Yahoo's data infrastructure. Additionally, the current Apache Hadoop market is forecasted to surpass $16 billion by 2020.
This might lead you to believe that Apache Hadoop is currently the backbone of data infrastructures for all enterprises; however, widespread enterprise adoption has been shockingly low.
While the platform is a key technology for gaining business insights from organizational Big Data, its penetration into enterprises has not lived up to Hadoop's game-changing business potential. In fact, according to Gartner, "Despite considerable hype and reported successes for early adopters, 54 percent of survey respondents report no plans to invest [in Hadoop] at this time, while only 18 percent have plans to invest in Hadoop over the next two years," said Nick Heudecker, research director at Gartner.
These findings demonstrate that although the open source platform may be proven and popular among seasoned developers who require a technology that can power large, complex applications, its fragmented ecosystem has caused enterprises difficulty extracting value from Apache Hadoop investments.
Another glaring barrier to adoption is the rapid and fragmented growth happening with Apache Hadoop components and its platform distribution, ultimately slowing Big Data ecosystem development and stunting enterprise implementation.
For legacy companies, platforms like Apache Hadoop seem daunting and risky. If these enterprises aren't able to initially identify the baseline business value they stand to gain from a technology, they are unlikely to invest - and this is where the value of industry standards comes into play.
Increasing adoption of Apache Hadoop, in my opinion, will require platform distributions to stop asking legacy corporations to technologically resemble Amazon, Twitter or Netflix. Through compatibility across platform distribution and application offerings for management and integration, widespread industry interoperability standards would allow Big Data application and solution providers to offer enterprises a guaranteed and official bare-minimum functionality and interoperability for their Apache Hadoop investments.
Additionally, this baseline of technological expectation will also benefit companies looking to differentiate their offerings. Similarly, standards within this open source-based Big Data technology will enable application developers and enterprises to more easily build data-driven applications - including standardizing the commodity work of the components of an Apache Hadoop platform distribution to spur the creation of more applications, which boosts the entire ecosystem.
A real world illustration of standardization in practice occurs within the container shipping industry, which was able to grow significantly once universal guidelines were implemented. When a formal shipping container standard was implemented by the International Standards Organization (ISO), to ensure the safe and efficient transport of containers, its significant impact increased trade more than 790 percent over 20 years - an incredible case for the unification and optimization of an entire ecosystem to ensure its longevity.
To help today's growing enterprise buyer looking to harness the estimated 4ZB of data the world is generating, the open data community will need to work together to foster the support of standardization across Apache Hadoop to ensure confidence from new adopters in their investment - regardless of the industry they serve.
From platform distributions, to application and solution providers and system integrators, known standards in which to operate will not only help to sustain this piece of the Big Data ecosystem pie, but it will define how these pieces interoperate and integrate more simply for the benefit of the ever-important enterprise.
Oct. 28, 2016 07:30 PM EDT Reads: 1,489
Oct. 28, 2016 07:15 PM EDT Reads: 671
Oct. 28, 2016 07:00 PM EDT Reads: 386
Oct. 28, 2016 07:00 PM EDT Reads: 4,243
Oct. 28, 2016 07:00 PM EDT Reads: 402
Oct. 28, 2016 06:30 PM EDT Reads: 376
Oct. 28, 2016 06:30 PM EDT Reads: 662
Oct. 28, 2016 06:15 PM EDT Reads: 2,939
Oct. 28, 2016 05:45 PM EDT Reads: 1,727
Oct. 28, 2016 05:30 PM EDT Reads: 319
Oct. 28, 2016 05:30 PM EDT Reads: 3,331
Oct. 28, 2016 04:45 PM EDT Reads: 328
Oct. 28, 2016 04:30 PM EDT Reads: 1,230
Oct. 28, 2016 04:30 PM EDT Reads: 780
Oct. 28, 2016 04:30 PM EDT Reads: 5,256