The Rise Of ‘Adaptive’ Applications

Applications are changing. So much so that technology commentators use the term ‘adaptive applications’ to define apps that are in some way more changeable and chameleonic in their nature. But the jury is out on what this term really means.

In some circles, adaptive applications are almost organic (like us) in their form and structure i.e. they can grow, shrink, skew and defend themselves against cyber attacks and naturally expand and develop as new connection points are added over time. An alternative view is that adaptive application excellence comes from a service agility perspective i.e. if an IT system app in one location is suffering from application performance degradations, then the adaptive apps within that deployment can switch channels and work with alternative cloud services or data streams. Still others think that adaptive apps have inherent self-healing characteristics and can autonomously update and perform maintenance tasks.

With a variety of viewpoints to draw upon in this space, the safest best is to define any ephemeral technology by its stance on data.

Cloud database platform company Couchbase defines adaptive applications as those elements of enterprise software that can straddle both ‘transactional’ and ‘operational’ data through ‘analytical’ systems using real-time technologies. But those terms are potentially confusing (surely business transactions are part of business operations, right?), so what do they mean?

What is transactional vs. operational data?

Put simply, there are as they sound i.e. transactional data forms part of a business transaction such as a sale or purchase, operational data records are the higher level tiers of data that detail the state of the business itself such as stock inventory levels, employee staff figures, team schedules and working hours etc. Potentially attached to both transactional and operational systems are data analytics services, so effectively that’s a third realm for data to have to travel through, often now in real-time. This all means that there’s a level of complexity here.

According to Forrester Research, "Moving data from transactional systems to operational systems and then to analytical systems is a barrier to making better decisions faster. Disparate data stacks also compromise the delivery of timely data to various applications, operational systems and into the analytics workflow, resulting in missed business opportunities."

Couchbase is aiming to address these challenges with its Capella columnar service on Amazon Web Services (AWS). Capella columnar is a new service that introduces a columnar store (a method of story data in columns as opposed to rows, often associated with an ability to achieve higher performance) and data integration into the Capella Database-as-a-Service (DBaaS), thereby allowing for real-time data analysis on the same platform as operational workloads. By converging operational and real-time analytics applications into one database platform, Couchbase claims to be able to remove friction and pave the way to adaptive applications in the data-centric sense.

“To gain a competitive advantage and accelerate business opportunities, organizations are investing in real-time, adaptive applications that are contextual, hyper-personalized and intelligent. Real-time analytics is paramount to delivering these adaptive applications and AI is making them even better and more strategic,” said Scott Anderson, SVP of product management and business operations at Couchbase. “With columnar in our Capella DBaaS, organizations can build adaptive applications powered by real-time analytics in a single JSON-based platform.” - JSON (JavaScript Object Notation) is a text-based, human-readable data interchange format.

Eliminating the latency gap

Anderson says his firm is working to ‘eliminate the latency gap’ that has always existed between analytics and operational databases while making it easier for software application development teams to include the required real-time analytic measures into their adaptive applications.

“The acceleration of AI creates an even greater challenge because having disparate database platforms introduces complexity that confuses the models being trained to power the application. Capella columnar solves these problems as it allows users to perform operational and real-time analytical processing with performance and speed, all in one architected platform. Furthermore, both the application and the user benefit by experiencing no latency and receiving a more seamless, premium experience,” added Anderson.

Capella columnar works within a Capella-powered application to enable fast, schemaless ingestion without having to perform extract, transform and load (ETL) tasks. The service can distribute data from operational workloads to perform real-time analytics on operational data and then immediately influence application behavior with that information.

The central idea revolves around working with more types of data (something every organization wants these days) and being able to save costs (something every organization has always wanted) by virtue of the fact that this technology can converge operational and real-time analytics in one data platform. Couchbase says that this means organizations can avoid absorbing the cost of one database platform for operational workloads and another for near-real-time analytics.

Capella columnar uses AWS services as an integral part of its product architecture including Amazon Elastic Compute Cloud (Amazon EC2), Amazon Simple Storage Service (Amazon S3), Amazon Elastic Kubernetes Service (Amazon EKS), Amazon Managed Streaming for Apache Kafka (Amazon MSK) and AWS Secrets Manager.

The future of adaptive apps

This discussion is not necessarily intended to attempt to coin or classify adaptive applications as those capable of spanning multiple data topographies; the initially tabled suggestions that application adaptivity might primarily focus on an ability to skew or scale up or down, capabilities related to cyber defense and wider competencies connected to maintenance and autonomous updates all may well still stand. Perhaps the most adaptive applications of all with exhibit all those characteristics AND also the ability to work across hybrid data streams.

In a computing landscape where change is constant and continuous, the ability to engineer application adaptability into our IT systems (with global disruption across infection, inflation and invasion so prevalent), we need to be able to change course when the situation arises. As one great philosopher once said, “We cannot direct the wind, but we can adjust the sails.”