8 Big Trends in Big Data Analytics

When it comes to big data, the experts agree. Don’t bother waiting in line for the first business-friendly iterations. These solutions are arriving on an innovation curve that’s ahead of formal adoption, says Robert L. Mitchell in a Computerworld article featuring data specialists Bill Loconzolo and Dean Abbot.

To the Cloud!

There seems to be a cautious but steady shift away from processing data sets in clusters on physical machines to cloud processing. Hadoop and Amazon Redshift are early leaders in this endeavor. One big breakthrough, also by Hadoop, is the distributed resource manager turned general-purpose data operating system. This means that business can run queries from multiple workloads, making it the ultimate multi-purpose data hub. Another major development is the concept of data lakes. In practice, it turns traditional database theory upside down. Instead of designing the database before data is entered, a mass heap of data is loaded into a repository, and then views will be built into the data sets on an as-needed basis.

Analyze This

As analytics go, traditional machine learning algorithms are being superseded by cheap computational power that surpasses concerns of speed and energy. And the capabilities of SQL on Hadoop are steadily increasing, says Mitchell. Thanks to companies like Cloudera and IBM, a new style of iterative analytics is becoming popular. In addition to improvements in traditional SQL, NoSQL (Not Only SQL) open source databases have come to the fore as faster, more direct ways to analyze customer-sales relations outside of the relational model.

Popular and Experimental Trends

Then there’s the trend of “deep learning.” This approach turns computers into detectives that use techniques based on neural networking to recognize items and relationships in large sets of unstructured data. Lastly, in-memory databases like hybrid transactional/analytical processing (HTAP) are quite popular because of their speed, but many experts point out the impracticality of maintaining an in-house product that often costs more than managing in the cloud.

Read the full article at:

Show More

Leave a Reply


We use cookies on our website

We use cookies to give you the best user experience. Please confirm, if you accept our tracking cookies. You can also decline the tracking, so you can continue to visit our website without any data sent to third party services.