Karthik Ranganathan, co-founder and co-CEO of Yugabyte discusses architectural enhancements in Yugabyte’s latest release, benefits and risks CTOs should consider, future database architectures and more in this chat.Â
————
Hello Karthik, please share about yourself and your role at Yugabyte. Also, tell us how Yugabyte has evolved over the past few years.
I founded Yugabyte in 2016 with Kannan Muthukaruppan and Mikhail Bautin, with the aim of evolving transactional databases for a cloud native world.
We built YugabyteDB, an open source, distributed SQL database that simplifies cloud migration and data management because we believed that the trends we saw at Facebook, including microservices, containerization, need for scale and high availability, and geographic distribution, were relevant to all businesses. This is especially pertinent as businesses move their legacy on-premises architectures to cloud-native operations.
Also Read:Â A Comprehensive Guide to DDoS Protection Strategies for Modern Enterprises
Since inception, we’ve innovated and enhanced our offerings for customers across cybersecurity, financial markets, IoT, retail, e-commerce, and other verticals. YugabyteDB has become widely adopted, serving business-critical applications with SQL query flexibility, high performance, and cloud-native agility. This enables enterprises to focus on business growth instead of complex data infrastructure management.
Most recently, we announced significant architectural enhancements to YugabyteDB that fully evolves PostgreSQL into a distributed database for modern applications.
As a company, we also reaffirmed our commitment to open source, not only for altruistic reasons (and because it’s good for business), but because we believe it’s the best way to advance the most critical technology development of our time.
What are the most significant architectural enhancements in Yugabyte’s latest release?
PostgreSQL remains the most popular database among developers, but modern cloud-native applications struggle to run on Postgres due to its single-node architecture and limited resilience capabilities.
Our latest release delivers on our promise to enable lift-and-shift modernization of Postgres apps to a fully distributed database that delivers built-in resilience and on-demand scalability.
As organizations respond to changing market conditions and customer demands, cloud-native applications are being rapidly adopted to improve business and operational agility. However, many of these applications were not built on databases designed to meet the demands of a cloud-native world.
YugabyteDB is now the only database that is both fully distributed (high data durability, availability, scalability, DR, multi-region) and runtime Postgres compatible. Newly released capabilities enable a broader range of Postgres apps to run on YugabyteDB. This makes it the perfect database for companies building cloud-native applications and modernizing existing applications to a cloud-native RDBMS.
With increasing regulatory scrutiny around data sovereignty and privacy, how can organizations ensure compliance while maintaining the agility and scalability of their distributed systems?
 With the proliferation of AI adoption, regulatory compliance has become an even greater requirement and priority for enterprises. To ensure compliance while maintaining agility, enterprises must stay current on the latest regulations in their regions and industries. It’s important to understand how laws around data privacy and access can impact operations. New AI tools can help ensure compliance, but human teams should be involved in overseeing data sovereignty and privacy practices.
To meet global customer experience and data privacy requirements, organizations are deploying globally distributed applications that allow data to reside in different countries. Geo-distributed databases allow organizations can meet the constraints of their data management strategy and easily and quickly configure their databases to meet the data sovereignty regulations of the host country.
By adopting a geo-distributed environment, organizations can meet agility and scalability objectives with synchronous and asynchronous data replication and granular geo-partitioning capabilities. This ensures compliance with local data privacy regulations, while delivering resilience, consistency, and low-latency across distributed systems.
What are the key benefits and risks CTOs should consider when adopting open-source technologies for critical infrastructure?
The open source community offers enterprises many benefits, including access to technologies that they may not have the resources or time to develop alone. Since this technology is backed by a community of engineers and developers, innovation happens faster, there’s more flexibility, and it is often the cost-efficient option.
No open source solution is perfect and you’ll need to take care to choose one that provides the capabilities your enterprise needs. You may have to compromise or work with the vendor to achieve this. It is a good idea to choose technology that is widely adopted and contributed to, to ensure you have community support and assistance if help is needed.
Companies operating in heavily regulated industries such as banking or insurance are often averse to open source technologies, perceiving them as more ‘risky.’ However, as the code of open source software is available for public review, security vulnerabilities are often found and corrected faster than proprietary software. There is also less rush to get untried features out of the door.
It’s also worth noting that although a vast range of technologies claim to be open source, many are capable of locking in customers to their offering. Consider the future direction of your open source software provider, as many enterprises have been stung by their open source software suddenly becoming closed (or hamstrung). Ultimately, you should do your research and read the fine print before selecting a business-critical open source technology.
Five years ago, we decided to make YugabyteDB fully open source under the Apache 2.0 license. We remain committed to open source. This decision is not only for business reasons, but also because we believe it provides a better experience for our customers. Customers help drive features and innovation, enabling us to grow with them as their database requirements evolve.
Also Read:Â Protecting APIs at the Edge
The nature of data workloads is rapidly changing, with real-time analytics, IoT, and AI. How do you envision the future database architectures catering to these new types of workloads?
Data has never been more critical to an organization’s success. Every transaction and (nearly) every decision is based on data. This means that it is important for organizations to implement a resilient and reliable database architecture.
Real-time analytics, IoT, and AI will only increase the strain on database architectures. For example, production deployments for activities like customer or operational transactions, which are popular use cases for YugabyteDB, must be available 24×7, so finding a database that can support these business cases is crucial.
Real-time analytics and data that fuel AI agents require similar uptime. If AI agents are making autonomous decisions, any latency or downtime in the database architecture could lead to delayed results at best and a catastrophe at worst.
Future database architectures must be built to scale and evolve with the growing needs of these emerging technologies. Focusing on resilience, performance, scale, and compatibility with the existing data stack will help ensure success and prevent organizations from having to retro-fit their database architecture every few years.
Why was PostgreSQL compatibility made a core feature of YugabyteDB, and how has it influenced enterprise adoption?
PostgreSQL is robust, fully extensible, and 100% open source. It is supported by a passionate community of developers—not to mention an expansive ecosystem of tools and frameworks. Over the years, Postgres has become the default API for cloud-native transactional databases.
Despite this, PostgreSQL has limited resilience and scalability. This is challenging for apps that need to stay online through planned or unplanned outages, scale quickly to meet demand, or keep data in multiple regions.
YugabyteDB was one of the first cloud-native databases to announce Postgres compatibility – as early as 2017-2018. We built a distributed database that reuses the PostgreSQL query layer so it looks and behaves just like Postgres. YugabyteDB retains all the power and familiarity of PostgreSQL while evolving it to an enterprise-grade distributed database with built-in resilience, dynamic scalability, and multi-site distribution. In essence, we built a Postgres-compatible, distributed database.
But we didn’t stop there.
A distributed database shards and replicates data to multiple nodes in a cluster to achieve resilience and scalability. When an application is ported from Postgres to YugabyteDB, query performance can be impacted, sometimes in unexpected ways. Developers had to spend extra cycles optimizing their existing Postgres apps for YugabyteDB’s distributed nature. Developer time is valuable, and while some enjoyed diving into distributed systems, we saw this friction as an opportunity for improvement.
We wanted to provide a better lift-and-shift experience, allowing developers to run their existing Postgres apps on YugabyteDB without additional work and being able to scale when they needed to without migrating to a different database.
YugabyteDB balances workload portability from PostgreSQL with scalability on demand when needed. Our mission with YugabyteDB has been to meet developers where they are so they can maintain familiarity and comfort with Postgres, but they don’t have to worry about changes or re-architecting issues as they scale.
Finally, what are your predictions for the future of distributed databases and their role in supporting AI and machine learning initiatives beyond 2024?
As enterprises enter the next phase of AI development and adoption, agility and seamless scalability will be essential. Gen AI tools and co-pilots enable developers to be more productive, allowing applications to be built faster, which will drive demand for a cloud native transactional RDBMS that can be deployed and scaled on demand in minutes.
Postgres-compatible distributed databases allow developers to start quickly, using a familiar interface and relational data modeling. Databases-as-a-service allow them to spin up a database in minutes.
Organizations are starting to build new applications and services powered by AI. These applications will require a low latency, highly available, scalable operational database to store data and to apply trained machine learning models to transactional data.
Business needs change quickly and frequently, so enterprises need to be able to act fast to meet customer demand and to drive value. Along with this, AI workloads are extremely resource-intensive, so teams are increasingly facing more data silos.
In 2025 and beyond, I expect teams to continue AI experimentation and work on addressing data silos to improve efficiency and performance. A distributed database will be crucial for enterprises trying to achieve AI agility. Distributed databases ensure scalability and consistency for AI/ML initiatives by eliminating silos and creating a stable, always-available environment for data management and access.
[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]
Karthik Ranganathan is co-founder and co-CEO of Yugabyte, the company behind YugabyteDB, the open-source, high-performance distributed PostgreSQL database. Karthik is a seasoned data expert and former Facebook engineer who founded Yugabyte alongside two of his Facebook colleagues to revolutionize distributed databases.
Yugabyte is the company behind YugabyteDB, the open-source, high-performance distributed SQL database for building global, cloud-native applications. YugabyteDB serves business-critical applications with SQL query flexibility, high performance, and cloud-native agility, thus allowing enterprises to focus on business growth instead of complex data infrastructure management. It is trusted by companies in cybersecurity, financial markets, IoT, retail, e-commerce, and other verticals. Founded in 2016 by former Facebook and Oracle engineers, Yugabyte is backed by Lightspeed Venture Partners, 8VC, Dell Technologies Capital, Sapphire Ventures, and others.
