APAC enterprises are migrating from proprietary databases to open-source alternatives, particularly PostgreSQL, due to high licensing costs and restrictive vendor dependencies.
As businesses accelerate digital transformation, IT leaders are rethinking database strategies to ensure companies meet evolving performance, cost, and scalability requirements.
“Three key trends are shaping database decisions globally. There is a movement toward open-source databases, particularly PostgreSQL, which has emerged as the de facto choice for enterprises, adopting it to modernise data infrastructure,” EnterpriseDB’s chief executive officer, Kevin Dallas told iTnews Asia.
He said enterprises are developing applications that require transactional, analytical, and generative AI workloads, making it crucial to adopt databases capable of handling all three.
At the same time, organisations are also adopting hybrid IT over pure cloud, balancing on-premises and cloud workloads to meet sovereign data regulations and AI governance requirements while ensuring control and compliance, he added.
These trends are particularly significant in APAC, where businesses are dealing with increasing transaction volumes and compliance needs.
EDB’s regional vice president and general manager, APJ, Stuart Fisher, said that countries including Singapore, Indonesia, and the Philippines are seeing strong adoption, driven by high transaction volumes.
“Japan and Korea maintain a strong on-premises demand, and India is leading the transition away from legacy database lock-in due to cost and ROI pressures.”
He added that enterprises are comfortable with multi-vendor environments and are consolidating around two or three key database technologies, and PostgreSQL is now recognised as one of them.
Role of PostgreSQL in hybrid IT
As hybrid IT becomes the dominant enterprise model, PostgreSQL needs to evolve to support emerging workloads, including AI and analytics.
To that end, Dallas mentioned that EDB has expanded PostgreSQL capabilities to meet the evolving needs of enterprises.
Dallas said one advancement is the adoption of a lakehouse architecture, which allows PostgreSQL to manage both structured and unstructured data.
This architecture enables enterprises to perform large-scale analytics on petabytes of data, making it a viable tool for AI-driven insights.
Another feature that enhances PostgreSQL’s AI capabilities is pgvector which enables translation between SQL-based queries and vectorised data - essential for applications involving AI-driven search, recommendation engines, and real-time decision-making.
This makes it possible to run AI applications directly within the database without requiring separate, specialised infrastructure.
According to Fisher, data sovereignty is a growing priority across the APAC region, with regulations mandating that data be stored and processed within national borders.
This poses challenges for cloud hyperscalers, as establishing local data centers in every country is not feasible.
“To address compliance challenges, enterprises are deploying PostgreSQL across hybrid cloud environments, balancing control with scalability,” Fisher added.
Dallas further highlights the growing complexity of managing multi-vendor IT environments.
To streamline security, governance, and compliance, the company has developed a hybrid control plane, offering a unified management interface across on-premises and cloud deployments, said Dallas.
He added that this centralised system enhances visibility, security, and regulatory compliance by enabling enterprises to monitor and manage the entire database ecosystem from a single platform.
Despite the advantages of PostgreSQL, enterprises often prefer a gradual transition. Legacy dependencies, staff expertise, and operational risks often make a phased approach to the more practical choice.
Full migrations are rare
Fisher mentioned how an insurance company in Hong Kong fully migrated from an Oracle proprietary database to PostgreSQL for all transactional workloads.
This has allowed the company to reduce licensing costs, increase flexibility, and improve performance while maintaining the security and compliance required in the highly regulated insurance industry.
“However, full migrations are rare. Instead, organisations start with specific use cases, such as microservices or new digital services, before gradually transitioning legacy workloads,” Fisher said.
Data volumes are increasing across industries including retail, shipping, manufacturing, banking, and taxation, where real-time processing of high-volume transactions is essential.
While incumbent vendors have traditionally served these environments, Fisher emphasises that PostgreSQL, with enterprise-grade enhancements, now meets scalability demands.
Adding on, Dallas mentioned that successful migration requires addressing two key aspects: database mobility and application portability.
While moving the database is a significant step, ensuring applications run efficiently on PostgreSQL is equally important.
Talking about future plans, Dallas mentioned that EDB is integrating GPU support and building tools to help customers accelerate the development of AI-powered applications.
Businesses today need to create and deploy generative AI solutions faster, what once took months should now take days.
“To achieve this, we are investing in tools that will allow users to seamlessly integrate their proprietary data, choose the right language models, and deploy AI agents efficiently,” said Dallas.
These capabilities will roll out progressively, with NVIDIA GPU support and other enhancements expected by the second quarter of 2025, he added.
