The evolution of the NoSQL database as a platform for real-time analytics, edge computing and its capacity for scalability, high performance, and data model flexibility has caught the industry’s eye. The NoSQL market is packed with heavyweights such as Redis, MongoDB, IBM Cloudant, Azure Cosmos DB, Cassandra and many others. With the rise of digital natives, future growth of NoSQL market will come from its applications in web session management, user profile management and e-commerce support across industries.
In an interview with Toolbox, Ravi MayuramOpens a new window , CTO of Couchbase, talks why NoSQL is now a preferred choice for businesses across industries, and why it is ideal for real-time analytics and edge computing. In fact, he predicts NoSQL will become the default database for all kinds of applications in five years.
Key Takeaways From This Tech Talk Interview:
- Emerging use cases of NoSQL
- Evolution of NoSQL to adapt to new challenges
- NoSQ leads the disruption of database-server-application architecture
- New trends in the database market for 2020
Here’s the edited transcript of the interview with Ravi Mayuram:
Tell us about your career path, your role at Couchbase and your technological direction which has made the company a strong leader in NoSQL database space?
Throughout my career I have built system software and platforms — operating systems, databases, search, application servers, mobile platforms, PaaS and distributed systems. Along the way I went from being a deep-in-the-weeds developer to senior management positions in the Engineering/R&D career track.
I joined Couchbase almost seven years ago, when it was a simple cache/key-value store — one of the best performing and one with the cleanest architecture. It has been a tremendous journey for us here at Couchbase where we built on that solid architecture and brought it to be a complete database and along the way tremendously advance the definition of a database.
Today, Couchbase is powering mission-critical applications for large enterprises. In your view, how has NoSQL database technology evolved?
When we started, we said we can handle your load, but you can’t have joins or ACID transactions and you have to use a bizarre proprietary query language. Some of the early NoSQL databases even today barely scale or provide much more availability than their relational predecessors.
NoSQL databases have evolved beyond that brittle database you use for only your high load use case where you have no other choice and have become more generally useful. Couchbase, for example, scales to thousands of nodes, provides joins, distributed ACID transactions and despite the “NoSQL†moniker actually supports SQL (because “NoSQL†means “not only SQLâ€).
When we started, NoSQL databases were almost exclusively operational systems, but now there are analytical NoSQL databases and some like Couchbase that are hybridized and provide an embedded MPP system for analyticsOpens a new window .
This is not your older brother’s NoSQL database. For many companies from the people that handle your credit cards to the ones that deliver your pizza and even the one that rents you a room for the night, this is their core database technology. As we move ever more to the cloud and now to edge computingOpens a new window technologies NoSQL databases are continuing to evolve.
Learn more: The Next Frontier for Big Data: Why Data Analytics is Moving Away from Centralized SystemsOpens a new window
There’s a cohort of companies around NoSQL database infrastructure – Redis, MongoDB, Aerospike. How’s Couchbase differentiating its product and solutions stack from its competitors?
Couchbase is the only database that combines the best of NoSQL with the ease and power of SQL, all in one platform. With key value store and document database capabilities, Couchbase is a multi-model database whose point of differentiation is within business-critical applications in the enterprise. And where we are uniquely positioned is in our ability to scale, perform at scale, and do so reliably.
Thanks to our unique in-memory, shared nothing, scale out, distributed architecture, Couchbase is routinely used for caching layers, sources of truth, and systems of record across high-scale as well as high-flexibility use cases, including offline-first applications at the edge. By design, Couchbase is accessed (SQL/SQL++ for JSON, full-text search, analytics) and managed through a consistent set of APIsOpens a new window , and scaled, upgraded, and diagnosed as a single unit, making Couchbase a complete database platform that not only addresses the needs of today, but offers the flexibility to adapt to the needs of tomorrow.
And with cross datacenter replication (XDCR) technology, as well as the Couchbase Autonomous Operator for KubernetesOpens a new window , Couchbase is cloud neutral, providing true cloud interoperability to give customers the freedom to pick any cloud. What all this means for our customers is they can seamlessly migrate from relational to NoSQL: no need to disrupt their development environment, no need to rip and replace. Businesses can be confident that their applications will deliver uncompromised performance and availability, at any scale, and at any location, now and in the future.
Learn more: Huawei, Aiming at Database Market, Takes on Oracle, IBMOpens a new window
Mobile first companies are tapping into the power of NoSQL because they have this pressing need for scale -high performance, agile processing of information at scale. Can you share some recent use cases from your market-leading customers?
Mobile first companies are tapping into the power of NoSQL because they have this pressing need for elastic scalability and guaranteed high availability of data and fast response times regardless of the state of the network. A NoSQL embedded database eliminates the need for expensive data migrations that reduces app startup cost and aligns well with object-oriented programing paradigm used in mobile applications.
One use case in particular, with BDOpens a new window , a leading global medical technology company, addresses a major obstacle to optimized treatment of Type 2 diabetes: the lack of data. BD’s solution is a mobile patient app and clinician portal built on Couchbase and Couchbase Mobile. The solution connects medical devices with a mobile app to automatically log real-time data on patients and give them custom alerts and recommendations. Doctors also use the data to determine the best course of treatment. Couchbase enables the app to work offline and securely syncs patient data from device to cloud.
How has Couchbase’s customer base diversified over a period of time? Has the company entered new sectors with a new set of use cases?
We’ve had quite a reach for years from our base in the travel industry to financial giants and of course retail. While Couchbase now reaches across industries, the big story is that we’re reaching deeper into our customer accounts. Some of our competitors have a great logo, but when you look at it, their customer uses their technology for a single use case. Couchbase has become trusted for our customers throughout their organization. We’re no longer just the database you use for a high volume customer site, but the database that you trust for all of your back-office needs.
And to be honest, I don’t think this is a NoSQL trend, it is a Couchbase trend. The thing that makes Couchbase special is that it can truly scale to meet your needs, but it also supports SQL and thus your ecosystem of other tools. It supports your developers who love to be able to write JSON and use a flexible schema, but it also supports your analysts with a built-in MPP and support for SQL++.
The depth of our technology is the difference, with our analytics support we’re seeing companies use Couchbase where in the past they might have used Teradata or Hadoop. I think that our newly announced Couchbase Cloud will be an accelerator for all of this.
The next big wave of Couchbase adoption will come from the combination of our support for truly reactive computing via our event engine and from our edge computing support. With the major mobile providers deploying massive edge networks, anyone who wants to take advantage of that is going to have to have a database that has a production-ready mobile to edge to cloud solution. And polling for data is just not an efficient option, they need a database that can react. Couchbase stands alone when it comes to a mature already in production solution for reactive, mobile and edge computing.
Learn more: Challenges to Enterprise-wide Adoption of Big Data TechnologiesOpens a new window
Has the market for NoSQL benefitted from Hadoop’s demise, the batch-oriented system which failed to catch up with the operational needs of internet-first companies? Can NoSQL fit into batch-oriented use cases?
HadoopOpens a new window was a technology designed for an earlier era that just missed the mark of current business and technology trends. I believe that business has been moving to what we used to call ubiquitous computing but now we call mobile and IoT. Meanwhile businesses have also been moving to “real-time†— that’s both cultural and technological.
Not long ago everything was done in person. Computers were there to help calculate things and data was generally transferred on paper. There was a lot of latency in the way business operated. Mail-order catalogs that got you products in 6-8 weeks were “normal.†When you’re on a 6-8 weeks for delivery timeline, data can be batched. You can wait for weeks or even months to know about your product return rate. Was your marketing successful? You’ll know at the end of the quarter.
When I click “order now†on my phone, I’ve come to expect same-day or next-day delivery. Now businesses need to know “now†about their KPIs. They need to provide customers data in “real-time†and be able to analyze their data in “real-time.†There just are fewer and fewer use cases where batch is acceptable. Real-time analytics are no longer nice to have but essential.
Can I use NoSQL in a batch use case? Sure, and people do that every day. We have customers who pull back data and do predictive marketing and offer analytics (i.e. which “add on†products or offers might result in higher margins) — for instance. However, while technically this is done as a batch, it is done on a much faster timeline than traditional batch technologies like Hadoop.
NoSQL is moving in on analytics because in a real-time world you need flexible real-time analytics. Even if some are batch, they’re done on shorter timelines than the companies of yore that delivered in 6-8 weeks.
Learn more: When should you switch from batch to streaming-based data processing?Opens a new window
The DBMS market has heavily evolved – and legacy companies like Oracle and IBM are also redefining their strategy. Against this backdrop, can you share how NoSQL players like Couchbase are staving off competition from entrenched market players?
In 5 years NoSQL will be replaced as a term. We won’t say NoSQL, we’ll just say “database.†We’re becoming the “default†for new applications. However, let’s dispense with any idea that companies are ripping and replacing on any kind of mass scale. It is happening in pieces. Relational databases won’t disappear just like Mainframes haven’t disappeared. However, when companies started creating their e-commerce or web-presence — they might have started with some link to their mainframe but it didn’t scale and the quickly moved to microcomputers. Against this backdrop of migration to the web as a primary way of doing business, there was a migration from the mainframe.
With the explosion of data and mobile technologies and the move to the cloud, companies are also going to move to cloud-native databases that scale and have the uptime required. Meanwhile, application deployment times are changing. Companies are pushing changes every week or even every day, NoSQL will achieve primacy in any company moving that quickly.
Legacy players tend to produce offerings that are intended to mark a place in a new market without damaging their flagship product’s margin or market share. Our strategy is to align our technology with what customers need and the business pressures they are under. We have no legacy product that will be challenged by this.
Learn more: Hadoop or NoSQL? What is the difference?Opens a new window
In closing, can you share trends that will shape the database market in 2020.
While we move to the cloud, we’ll start to move to the edge. Customers expect absolute uptime. They expect next to no latency. Moreover, everything will become even more realtime. This means the difference between the database and the application will start to disappear.
Event-based technologies and edge computing are going to become hot topics. The architecture these changes suggest differs substantially from the standard three-tier client-application server-database architectureOpens a new window of today. The database will move into the client, the edge and yes the cloud. The application will move more into the database or the database will move more into the application.
Learn more: 6 Ways to Prep Your Database for a Microservices ArchitectureOpens a new window
What are some of the fast-growing markets where Couchbase wants to establish its presence as the leading database infrastructure provider in the next 2-3 years?
There are so many industries being transformed — travel, where we are used globally on land (Marriott), air (United), and sea (Princess Cruise Lines); or finance, where we are used by various financial institutions; or manufacturing, because all of these industries are transforming, moving to the cloud, and adopting new technology.
Instead, I’d point out that Couchbase has been busy building a global organization from EMEA to Asia Pacific. We’re already seeing returns on that investment. I expect that we’ll continue to see outsized growth in those global markets even as we continue our success in the US.
About Ravi MayuramOpens a new window :
As Senior Vice President of Engineering and CTO, Ravi is responsible for product development & delivery of the Couchbase Data Platform, which includes Couchbase Server and Couchbase Mobile. He came to Couchbase from Oracle, where he served as senior director of engineering and led innovation in the areas of recommender systems and social graph, search and analytics, and lightweight client frameworks. Also while at Oracle, Ravi was responsible for kickstarting the cloud collaboration platform. Previously in his career, Ravi held senior technical and management positions at BEA, Siebel, Informix, HP, and startup BroadBand Office. Ravi holds a Master of Science degree in Mathematics from University of Delhi.
About CouchbaseOpens a new window :
Unlike other NoSQL databases, Couchbase provides an enterprise-class, multicloud to edge database that offers the robust capabilities required for business-critical applications on a highly scalable and available platform. As a distributed cloud-native database, Couchbase runs in modern dynamic environments and on any cloud, either customer-managed or fully managed as-a-service. Couchbase is built on open standards, combining the best of NoSQL with the power and familiarity of SQL, to simplify the transition from mainframe and relational databases.
Couchbase offers NoSQL solutions to industry leaders Amadeus, American Express, Carrefour, Cisco, Comcast/Sky, Disney, eBay, LinkedIn, Marriott, Tesco, Tommy Hilfiger, United, Verizon, as well as hundreds of other household names.
About Tech TalkOpens a new window :
Tech Talk is a Toolbox Interview Series with notable CTOs and senior executives from around the world. Join us to share your insights and research on where technology and data are heading in the future. This interview series focuses on integrated solutions, research and best practices in the day-to-day work of the tech world.
Do you think the adoption and use cases of NoSQL databases will rise in the future? Comment below or let us know on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We’d love to hear from you!