Data is at the center of modern applications, placing new demands on database technology. But building distributed databases to complement distributed applications has proven difficult.

SiliconANGLE Media’s livestreaming studio theCUBE went deep with experts during the recent Couchbase ConnectONLINE event, exploring tough questions around the topics of SQL vs. NoSQL, to move or not to move edge data, and why data lakes turned to data swamps.

TheCUBE’s coverage of Couchbase ConnectONLINE featured interviews with company executives, technologists and end users who hashed out the latest developments from the database world. Here are five insights you might have missed from our in-depth conference coverage. (* Disclosure below.)

1. The debate over SQL vs. NoSQL is over — both win

The debate over relational, or SQL, versus NoSQL database has raged for years. Most recently, NoSQL databases have taken the lead in modern IT shops. But much about this debate is misunderstood, beginning with the terminology, according to Ravi Mayuram, senior vice president of engineering and chief technology officer at Couchbase Inc. 

While databases labelled “NoSQL” improved on relational databases, in some ways those improvement didn’t come from eliminating SQL (Structured Query Language), according to Mayuram.

“NoSQL was the best approximation and a good marketing name. But from a technologist’s perspective, the problem was never the query language,” he said in an interview with theCUBE. “SQL has been the language that has talked to data for 40-odd years successfully. Every other database has tried to implement their own QL query language, but they’ve all failed.”

There are unavoidable mathematical reasons for this, he added. In practical terms, SQL lets the user query data, in plain language, without having to move, transform or recast it. This provides “optionality” within a single data set, Mayuram explained. 

The real problems with relational databases: schema limitations and the inability to easily scale horizontally. Couchbase Server 7.0 — released last July — eliminated these problems and retained SQL to deliver cloud database technology for modern, distributed applications, Mayuram pointed out. 

2. Petabytes of data won’t fly; moving logic to data solves distributed DB problems

Data is distributed by nature, but building a distributed database has proven a hairy challenge for technologists. Distributed computing and applications, and the need for data at the edge, are creating a gaping need for a solution. What tech has done wrong for the last four decades, according to Mayuram, is insist on writing schemas before writing a single byte of data to the database. Couchbase has turned this on its head by writing data first, which then “late-binds” to schema on read. 

Couchbase houses and manages data in its native form of JSON documents. Its “optionality” enables key-value lookups, full SQL querying, and search with a full textual analysis pipeline. It also allows users to write their own custom logic on it with “eventing” capabilities. 

Moving petabytes of data around distributed systems and edge points presents a humongous challenge. Bringing the logic to the data instead is the way forward for distributed databases and applications, according to Mayuram. 

“I believe a database which gives you this optionality of doing multiple data processing on the same set of data will allow you to build a class of products which you have been struggling to build,” Mayuram told theCUBE. 

3. Couchbase aims to make good on broken promise of data lakes

Remember the buzz about data lakes — the vast repositories of data without schemas? They eventually turned into “data swamps,” leaving data scientists and engineers cleansing and preparing data 50-80% of their time. So much for their promise of instant insight and innovation at enterprises’ fingertips. 

How do we know Couchbase’s cloud database as a service — essentially a huge collection of schema-free data in JSON document form — won’t impose the same problems on users? 

For one, data swamps were a bit too open and also tossed in data in all sorts of forms. This resulted in poor usability and arduous integration duties. Sticking to JSON documents, imposing schemas later and adding a management layer negate these problems, Couchbase claims.

The management and automation features of Capella — based on Couchbase Server 7.0 —  shift the focus of companies away from data grunt work to real innovation and business use cases, according to Mary Roth, vice president of engineering operations at Couchbase.

“I started my career as a DBA, and it was one of the most sought after and highly paid positions in IT because operating a database required so much work. [But] with Capella, what we’re seeing is it’s taking that job away,” Roth said in an interview with theCUBE. 

4. Hosted versions, expansion as a public company, top Couchbase’s 2022 agenda

Couchbase now offers a hosted version of its Capella database as-a-service offering. The hosted offering runs within the Amazon Web Services Inc. cloud and includes the usual Couchbase infrastructure and management.

The hosted version offers a more convenient onramp to Capella, enabling developers to start building applications right away. And, since it runs in Couchbase’s AWS account, users no longer need to order their own AWS Virtual Private Cloud to run Couchbase. Hosted versions for Microsoft Azure and Google Cloud Platform are likely to debut 2022.

Such offerings may help Couchbase acquire more enterprise customers as it seeks to expand as a newly public company. It recently reported over $100 million in revenue and now boasts over 30% of the Fortune 100 as users. The NoSQL market as a whole is projected to grow to $22 million by 2026, up from 2,410.5 million in 2018, according to an Allied Market Research study published in early 2020. 

“Enterprises want freedom of choice in how they consume, deploy, run and manage their database,” Matt Cain, president and chief executive officer of Couchbase, told theCUBE. “They want the fastest path to developer agility and productivity, and they want the best total cost of ownership relative to other database-as-a-service offerings. That is exactly what we have provided with Couchbase Capella.”

5. Advances in distributed DB pop open can of new edge apps

The explosion of the mobile app economy is due in part to capabilities near and dear to Couchbase. They include a single, central location for data and the option for multiple systems and/or processes to converge on that data at once. This is why Couchbase advocates a model that does not ask data to disembark from where it is — on a phone, in an car, a home or industrial internet of things device, etc. — but instead brings logic to that data. This, the company believes, will supercharge distributed and edge app development. 

For example, Couchbase customer PepsiCo Inc. built an mobile app for its sales force that collects data — even offline — on a Couchbase database on the device. It then synchs the data as soon as a connection is established.

“It’s a very complex system, and from a volume perspective it’s huge. Couchbase takes care of the synch, and then we can focus on our business process,” Madhav Mekala, PepsiCo principal architect, said in an interview with theCUBE. 

Cain underscored the practical app focus of its database as-a-service offering in a separate interview with theCUBE.

“We support the SQL query language, we’ve got operational analytics, we’ve embedded full text search, and we have things like eventing. All of these are elegantly architected features that allow developers to build great applications,” Cain said. 

Watch Cain’s exclusive segment for theCUBE’s coverage of Couchbase Connect below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of the Couchbase ConnectONLINE event.

(* Disclosure: TheCUBE is a paid media partner for the Couchbase ConnectONLINE event. Neither Couchbase Inc., the sponsor for theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE