While advancements in techniques and technologies to capture, store, retrieve and analyze new forms of unstructured data have grabbed headlines, most organizations continue to focus primarily on managing structured data, and will do so for the foreseeable future, according to a new study conducted by Quest Software and Unisphere Research.
The global survey was conducted to explore the latest trends in data management, as well as gain insight into the adoption rates of Hadoop, NoSQL, and other modern database management technologies. The survey drew responses from 300 database administrators and others charged with managing corporate data from a wide range of organizations in terms of size and industry.
Unstructured data types and new database management systems are playing an increasing role in the modern data ecosystem, but structured data in relational database management systems (RDBMSs) remains the foundation of the information infrastructure in most companies. In fact, structured data still makes up 75% of data under management for more than two-thirds of organizations, with nearly one-third of organizations not yet actively managing unstructured data at all.
According to John Whittaker, executive director, product marketing, Information Management, Dell Software, there were three salient points that stood out from the research:
Role of RDBMSs Remains Strong
Though the survey reveals a relative diversity of database platforms in use across organizations, the majority of respondents indicated that they were running mission-critical data on Oracle and Microsoft, followed by MySQL, IBM DB2, and MongoDB as the next most popular database management systems.
To download a copy of the report, “The Real World of the Database Administrator,” go here.
Moreover, although the growth of unstructured data has garnered attention, Dell’s survey shows structured data growing at an even faster rate. While more than one-third of respondents indicated that structured data is growing at a rate of 25% or more annually, fewer than 30% of respondents said the same about their unstructured data.
“It is not a scenario where Hadoop and NoSQL are sidelining relational databases. It is not that at all,” said Whittaker. “If anything, the indication is that relational databases are continuing to grow and in fact 83% of organizations cited transactional data as one of the most important sources of structured data growth in their organizations, and the second ERP systems that are continuing to grow. The apps and the transactional data that predominantly run on relational databases are growing.”
And, although there is an increasing industry focus on the proliferation of social data, an increase in the creation of internally generated documents was seen as the top driver of unstructured data growth, identified by more than 50% of respondents.
Although respondents indicated interest in adding more database management systems to achieve added benefits, the adoption of technologies such as NoSQL and Hadoop is not yet truly widespread. Only 10% of respondents mentioned they are currently using or deploying a NoSQL database, while 56% of respondents claim their companies do not have plans to adopt one within the next 3 years. The results are similar for Hadoop. Approximately 20% of organizations surveyed are currently using or deploying Hadoop, with 57% indicating their companies have no plans to incorporate Hadoop technology within the next 3 years.
However, the survey does provide indicators that more widespread adoption of these newer platforms may in fact be coming soon. The need to support new analytical use cases, which increasingly involves unstructured data and big data technologies, was cited as the most important factor driving adoption of new database management systems, with the need for greater flexibility and performance closely behind.
The appearance of MongoDB as one of the five most commonly used systems indicates a growing acceptance of NoSQL technology, particularly among larger organizations. Roughly 70% of of respondents using MongoDB are running more than 100 databases, 30% are running more than 500 databases, and nearly 60% work for companies with more than 5,000 employees. Similarly, 60% of respondents currently using Hadoop are running more than 100 databases, 45% are running more than 500 databases, and approximately two-thirds work for companies with more than 1,000 employees.
With this heterogeneity, organizations should look toward solutions, capabilities, and providers that have a broad perspective beyond just relational or NoSQL/Hadoop in order to maintain flexibility, said Whittaker. There is going to be an expansion of the heterogeneous world where organizations have multiple technologies in their environment. Each of the technologies offers benefits and that is why organizations are continuing to deploy both, he said.
“From Dell’s perspective,” he said, “it aligns with the strategy that we adopted as Dell Software emerged 3 years ago to take an “all-data” approach where we built our data management, data integration, and data analysis capabilities with products like Toad, Boomi, and Statistica. All are designed to support a heterogeneous data environment.”
DBAs Being Tapped for Relational and Non-Relational Systems Oversight
As modern information infrastructure continues to evolve, a growing number of DBAs are now responsible for managing both relational and non-relational database technologies, a trend that will likely continue as Hadoop and NoSQL become more common in the enterprise. According to the survey, among respondents in companies with both Hadoop and NoSQL installed, DBAs are responsible for managing the non-relational technologies 72% of the time. Roughly 70% of the respondents said the DBAs in their organizations managed databases from at least two vendors, while 7% indicated their DBAs were responsible for managing databases from five or more vendors. Almost half of the DBAs in the study manage more than 25 database instances each, and almost 10% manage more than 100 database instances each.
Increasingly, the survey shows that the DBA is bridging the world of the different data types, rather than a new role emerging, observed Whittaker. “Certainly, the data scientist is a new role and a great aspiration for DBAs who might want head down that path,” he said, but increasingly DBAs are simply becoming “multi-lingual” in order to support Oracle and Microsoft SQL Server, and as well as MongoDB or Hadoop in their environments.
In essence, the role of the DBA is broadening and orienting around the data, wherever it resides, he said.
To download a copy of the report, “The Real World of the Database Administrator,” sponsored by Dell Software and produced by Unisphere Research, a division of Information Today, Inc., go here.