NoSQL databases, in general, are well-suited to big data environments. “NoSQL databases have high scalability which makes them suitable for dealing with large volumes of data while offering cost-effectiveness,” added Arthur Iinuma, president of ISBX. “They combine management of unstructured data, flexible systems, and complex analysis. NoSQL databases also have high agility and dynamic schema, which makes them ideal for big data and Internet of Things applications.”
Open source databases also provide high degrees of flexibility and cost-effective performance. The variety and flexibility these next-generation databases provide is impressive, enabling “far more options than ever for engineers to match specs to their use cases. Further, these next-generation databases provide a stronger link between data and compute tools, such as Spark SQL, Presto, or Dremio,” Bloomberg’s LaPorte said.
In addition, this new breed of databases has “the ability to scale horizontally with more ease than ever before,” LaPorte continued. “For instance, Cassandra allows for doubling in size on commodity hardware with minimal effort from either a service provider or our in-house application teams.” Illustrating the performance and cost balance, LaPorte related how one of Bloomberg’s clients “reduced their hardware footprint by 50% by moving from a traditional RDBMS to Cassandra. As a side effect, they were able to increase both performance and future scalability.”
Stacy Robin, founder of The Disruptive Diva, points to the emerging class of AI databases that deliver advanced analytics as a basic feature. “Emerging technologies, independently, are useful. However, their real value is when you start creating mashups, she said. AI databases that are built for training machine learning and deep learning models are where businesses should be looking to truly transform the value of robust databases and take their intelligence a step further, Robin said.
WEIGHING THE RISKS
Of course, with any new wave of technology innovation, there are inherent risks that must be factored into designs and deployments. “As with traditional RDBMSs, not all use cases match next-generation databases,” said LaPorte. “When your business use case requires an ACID-compliant database, a traditional RDBMS may fit your use case better than current NoSQL implementations. The performance and scalability trade-offs provided by NoSQL technologies like HBase and Cassandra necessarily sacrifice these ACID guarantees.”
There are many considerations that need to be made when transitioning to next-generation database solutions, “including the capabilities of the future-state solution versus the current state, the impact to licensing and support contracts, and a method to ensure that the correct solutions are deployed,” said Syntax’s Caruso. “We are continuing to see use cases in which a suboptimal technology is chosen because the technical team wants to work with it.”