Tracking Service-Oriented and Web-Oriented Architecture

SOA & WOA Magazine

Subscribe to SOA & WOA Magazine: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get SOA & WOA Magazine: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


SOA & WOA Authors: TJ Randall, Lori MacVittie, Andreas Grabner, Dynatrace Blog, Cynthia Dunlop

Related Topics: Cloud Computing, SOA & WOA Magazine

Blog Post

Future Data Management Strategies & NoSQL By @MapR | @CloudExpo [#Cloud]

The rapid growth and adoption of NoSQL databases are a clear indication of their effectiveness in this new era of big data.

NoSQL databases are highly scalable and capable of solving a wide range of data problems, which make them great for supporting complex, unpredictable, and large volumes of data. While still considered relatively young in the enterprise software market--the current industry adoption is estimated at 20%--deployments can potentially double in 2017. This is reflective of the growth of emerging use cases that require scalable and flexible systems. In addition, organizations find that some of their applications currently deployed on relational database management systems (RDBMS) are better suited with the strengths of NoSQL systems.

RDBMS vs. NoSQL
An effective data management strategy must be executed with every data-related factor in mind: the nature of data sets, volume and velocity of data, scalability, and performance all play a role. Both major types of database management systems - RDBMS and NoSQL - have their own sets of ideal use cases. And while RDBMSs have been the foundation for enterprise architectures for more than 30 years, the importance of NoSQL systems has grown significantly in recent years. This is due to advantages like high performance, scalability, and relatively low cost. For your data management planning, it's worth the investment of time and energy to understand the latest NoSQL technologies, and how they compare with the RDBMSs you already have deployed. A broad knowledge base will help you prepare a successful data management strategy and leverage NoSQL technology in the most effective way.

Often, the most important consideration in the selection of a database is the nature of the data you want to leverage. If, for example, the data volume growth is very predictable--such as data that is entered by customer service representatives--with a known and fixed schema, then an RDBMS will suffice. This is why ERP, CRM, and HRMS systems are almost exclusively based on RDBMSs. However, RDBMS-based applications can suffer serious performance issues when faced with an extremely large data size and high number of users. RDBMS vendors typically recommend vertical scaling, but sometimes offer horizontal scaling (clustering) as a solution, but these options on RDBMSs are both expensive and complex in nature.

Here, NoSQL databases play an important role, as they have a strong foundation for supporting horizontal scaling (employing cost-effective commodity hardware), complex and/or changing data structures, sub-second response time, and high availability. Horizontal scaling, in which more data and users are accommodated by simply adding more nodes to a cluster, is ideal when data volumes grow so fast that it is hard to predict how much hardware to deploy in the near term. A NoSQL database supports schemaless data models, so they are a perfect fit for a database environment where one record's structure might differ from another. NoSQL databases in general simplify the transactional mechanism, and thus remove some unneeded coordination overhead, so they are capable of supporting millions of users without compromising on speed. Response time is swift, and data retrieval efficient.

NoSQL: the Answer to Complex Data Problems
It should come as no surprise that more and more companies are pursuing a NoSQL database management strategy. Since NoSQL databases are often used as a supplementary database management solution, especially for solving big data problems, you can expect to see more data centers with both technologies running side-by-side.

NoSQL systems represent numerous capabilities:

  • They support complex business applications, including operational, real-time, and predictive analytics.
  • They support fast and efficient data processing at scale, and scales horizontally across nodes in a clustered environment.
  • They support low-latency queries for mission-critical applications where the response time is crucial to success.
  • They can manage huge collections of structured, unstructured, and semi-structured data.
  • They support the changing characteristics of mobile, cloud, big data, and Web 2.0 applications.

NoSQL databases come in many forms; basic classifications relate to data models (e.g., document, wide-column, key-value, or graph). They do, however, represent one unified goal: the creation of a database platform to support the broader needs of enterprises and application development.

The rapid growth and adoption of NoSQL databases are a clear indication of their effectiveness in the era of ever-changing data requirements. NoSQL vendors are adding features right and left to support more types of workloads. Architectural innovations, like those by Hadoop and NoSQL vendor MapR Technologies, enable more use cases including those that require Hadoop analytics and require performance, reliability, and security. NoSQL is emerging as an independent database ecosystem with the potential to provide a comprehensive alternative to RDBMS when facing complex data problems that are not a fit for predefined, structured tables. As big data continues to be top of mind, NoSQL systems are here to take on the challenges.

To learn more about the top NoSQL technologies and how to best leverage them for your business, download The Forrester Wave: NoSQL Key-Value Databases report.

More Stories By Dale Kim

Dale is Director of Industry Solutions at MapR. His technical and managerial experience includes work with relational databases, as well as non-relational data in the areas of search, content management, and NoSQL. Dale holds an MBA from Santa Clara University, and a BA in Computer Science from the UC Berkeley.