Businesses today face an overwhelming “data tsunami” — a surge of data that, while valuable, often hampers application performance and strains architectural scalability. The more data an organization manages, the more challenging these issues become. With AI workloads adding to data management complexities, enterprises are looking for ways to achieve scalability, maintain performance, and meet evolving data demands.
CIO Dimension spoke with Bhanu Jamwal, Head of Presales & Solution Engineering, APAC, TiDB, to explore how a modern and innovative approach to databases is helping businesses overcome these challenges and innovate. TiDB – an open source distributed SQL database developed by PingCAP – tries to provide some answers.
How is PingCAP’s TiDB positioning itself in the competitive database market, particularly considering the rapid growth of digital native businesses in India?
We’ve witnessed a remarkable surge in digital native businesses across India, spanning e-commerce, logistics, fintech, gaming, SaaS, and more recently, the emerging healthcare and Gen AI sectors. This explosive growth, however, comes at a cost – a massive influx of data.
Their increasing reliance on data creates a greater demand for scalable and reliable data solutions. Traditional databases often struggle to scale to meet this kind of growth and volume. Imagine a scenario where an e-commerce platform, during a peak sales event experiencing a huge surge in transactions. Traditional databases, such as MySQL or PostgreSQL, may encounter performance bottlenecks under such pressure.
Businesses are therefore compelled to either implement workarounds or adopt modern database solutions that can address these challenges. And these challenges now go beyond just scale. High availability is paramount; downtime is simply not an option today. Real-time analytics is another critical factor businesses. For instance, a bank needs to be able to detect fraudulent transactions in real-time to safeguard customer funds. So, it needs, scalability, availability and analytics- all at the same time. AI introduces further complexities and challenges to this mix.
Recognizing these evolving needs, TiDB was born in 2015. We strive to address these challenges with our distributed architecture, which allows businesses to scale as their data grows. Distributed SQL databases have in fact emerged as an effective answer to modern data challenges. This approach has been gaining traction, as evidenced by Amazon’s recent entry into the distributed SQL market, acknowledging the growing demand for such solutions.
How do you view the SQL vs. NoSQL debate, and how does TiDB position itself in this context?
The SQL vs. NoSQL debate has been ongoing for some time with both playing pivotal roles in enterprise environments. While NoSQL databases offer flexibility for specific use cases, they often require significant development effort to maintain data structures. SQL, on the other hand, remains the de facto standard for data access, with a vast pool of skilled developers.
It’s interesting how many NoSQL databases, which originally distanced themselves from SQL, are now adding SQL layers to their platforms. It shows that SQL remains a simple and effective way to access data.
TiDB’s aim is to bring best of both by offering a distributed SQL solution. Our Hybrid Transactional/Analytical Processing (HTAP) capabilities are designed to bridge the gap between SQL and NoSQL databases and handle both transactional and analytical workloads.
How is AI and ML, particularly the rise of LLMs, changing the game for distributed databases, and how is TiDB capitalizing on these trends?
The rise of vector database is a trend worth mentioning in this context. Vector is becoming a powerful engine for today’s AI workloads. Though it has been around for a while, the adoption has rapidly scaled in recent years. These databases are becoming increasingly crucial for storing and processing the data generated by AI/ML models.
But, this also makes the environment quite complex for enterprises. They often manage multiple databases for different workloads—a transactional database, a separate one for analytics, and yet another for vector storage. Managing databases is a complex process.
At TiDB, we see an opportunity to simplify this process by offering a unified solution. Our vision is to provide a single database that handles all these workloads, eliminating the need for multiple systems. TiDB supports vector as a form of storage with TiDB serverless, enabling organizations to store and process both traditional data and vector data within a unified platform. We believe the integration of vector search capabilities into a traditional database will be a game-changer for businesses looking to leverage the power of AI and ML.
Can you share some successful case studies of TiDB implementations in India and globally?
Absolutely! We have a number of success stories across various industries. In India, a leading e-commerce giant was struggling to manage the increasing complexity of their database infrastructure as they scaled, relying on a multitude of single-node databases. By migrating to TiDB, they consolidated their infrastructure, significantly improved performance, and gained greater operational efficiency.
Delhivery, a leading logistics company that delivers a million packages a day, migrated to TiDB to power their real-time analytics and data marts use case, enabling them to handle thousands of requests per second with less than 100ms latency.
Pinterest is another flagship customer that leveraged TiDB to modernize their data architecture. Globally, we’ve seen tremendous success in the fintech sector. A major Japanese fintech company, for example, experienced a 10x increase in throughput and a 30% reduction in latency after adopting TiDB for their high-volume wallet application.
We’re also seeing significant growth in the SaaS market, where TiDB helps companies manage the complexities of multi-tenant environments and scale their databases to support growing user bases.
What are the key market opportunities for TiDB in India?
India is a dynamic and rapidly growing market with a thriving startup ecosystem. We see immense potential here. The sheer volume of data generated by India’s digital-native businesses and startups, coupled with their ambitious growth plans, creates a strong demand for distributed database solutions. India’s thriving Web3 community, its vast base of skilled developers are all opportunities that TiDB will be actively tapping into.
