Data Outliers: 10 Ways To Prevent Big Data Damage

With the need to capture and analyze data growing at a phenomenal rate, the global big data management solutions market is expected to soar at a compounded annual growth rate (CAGR) of 12.8% between now and 2021, according to a Forrester Research report.

Driving this growth will be non-relational databases and in-memory data fabric, Jennifer Adams, senior forecast analyst for Forrester, told InformationWeek on Wednesday. She noted the 12.8% growth rate for big data management solutions will also be faster than the rate for traditional data processing solutions.

"Forrester expects non-relational databases to be one of the fastest-growing sectors within big data management solutions. The complexity and richness of data is changing and unstructured data, such as text, tweets, graphs, and video, are an increasingly important source of information," Adams said.

{Image 1}

Non-relational databases are keys tools to manage and search these large and diverse data sets, and within the group NoSQL is forecasted to grow 25% and Hadoop 32.9% annually over the forecast period, she added.

Scaling is built into non-relational databases, in which millions of users and hundreds of terabytes of data can be supported, Adams stated in a blog post on Tuesday. In the post, she noted non-relational databases usually cost only 10% of what a comparable traditional relational database would run.

In assessing when non-relational databases like NoSQL make sense to use, often IT professionals are looking for flexibility and scalability. With NoSQL, an open source and horizontally scalable distributed computing architecture is used.

Hadoop's software ecosystem, meanwhile, provides a framework that allows users to crunch enormous amounts of data via a massive parallel computing architecture.

Companies are advised to consider taking a slow approach to stretching their data management capabilities, rather than deploying big changes to achieve improvements, according to a separate Forrester report provided to InformationWeek.

[See 9 Hot Big Data and Analytics Startups to Watch.]

One of the suggested steps is to "stretch the technical capability" of data management by enhancing its flexibility and elasticity. The report found that 61% of decision-makers characterized internal unstructured data as important or very important to their company's overall business strategy.

Some 35% cited externally available unstructured data as important. But, rather surprisingly, more than half of those same decision-makers who were surveyed were only using up to 25% of the available unstructured data to put it to use for insight into their business, according to the report.