By: Grace Chng

More than ever, digital transformation is essential for companies to survive and thrive. Due to the Covid-19 pandemic, brick-and-mortar business models are losing relevance as office workers, customers, and consumers, disappear from workplaces and shopfronts, sheltering themselves in the safety of their homes and preferring to interact virtually.

To survive, organisations have no choice but to digitally transform their business models to align themselves to the new normal. They need to be agile, flexible and innovative to pick up on unexpected opportunities that will emerge in the post-Covid era.

At a roundtable event, hosted by enterprise PostgreSQL provider and contributor, EDB, CIOs from across the Asia region discussed the pro’s and con’s of using open source databases. Following is a summary of the discussion.

For banks, open source software could be the key to digital transformation

Due to the COVID-19 pandemic, banks are grappling with a surge in online banking, which has created a pressing need for digital transformation. 

Behind the online or mobile bank app is traditional enterprise database software that holds the customer records and transaction details. With transaction volumes increasing, more data points are being created every second. While proprietary database software is a good repository of data, it is not that flexible for software developers to access datasets in order to create new services as a result of customer demand. Instead, open source databases offer features that help add flexibility and scalability to the software development process.

Silverlake Digital Economy, a provider of intelligent banking solutions in the ASEAN region, has demonstrated the ease with which it was able to deploy a digital core banking platform that was created on a PostgreSQL database. 

Silverlake piloted the Mobius platform with an ASEAN bank to test it for transaction speed and found that it could shorten the time to balance their books each day. 

Previously the Bank ran at 4,000 transactions per second, but Mobius could scale on demand to 18,000 transactions per second. Mobius can run on any cloud platform, without any hardware deployment. Previously, such a test would have necessitated creating an environment of hardware, operating systems and other components, requiring an investment in time and capital.  

Banks will traditionally shut down access to their mainframes each night so that programs can be run to automatically balance the books at the end of each banking day. For larger banks, it will typically take three to four hours to do this for several million accounts. With Mobius, it took a mere 90 minutes to process 20 million records.

Hong Yea Chee, Silverlake’s senior executive vice-president pointed out that the pilot demonstrates that new services can be quickly developed using the combination of open source software and cloud. 

“That’s why I think open source databases are essential to digital transformation,” Hong said, emphasising that with Mobius, innovation can occur faster because the open source database allowed for a quick sandbox to be set up in the cloud for experimentation without having to set up a physical infrastructure. 

With increasing data volume, open source helps with database load balancing 

Another compelling reason for organisations to adopt open source database software are new features like database load balancing. Increasingly, load balancing is becoming important as the volume of data collected by the Internet of Things is exploding. To better manage their data, organisations want to split up data to distribute and store it in smaller datasets so that it is easily accessible, which leaves the database more responsive to requests.

On top of this, such data processing activities are likely to be undertaken at the edge, leading to workload distribution across multiple databases in the cloud. 

“Any company with many customers and distributed workloads may often need to do this,” said Jacqueline Teo, Chief Digital Officer of HGC Global Communications, Hong Kong. With cloud, organisations do not have to worry about their application environment, because the databases move from one cloud to another as the backend software is usually powered by open source projects.

Then, there is the seamless integration of applications and services in the cloud, which makes databases convenient to manage and easy to access. Activities in the cloud made possible by open source database solutions include (with internet connectivity), access for remote workers to corporate information, virtual systems to connect with colleagues, engage with customers, and facilitate transactions.

Considerations for selecting open source databases

  1. Security

In selecting open source databases and moving workloads to the cloud, considerations should be made to an organisation’s perspective on security, said the senior ICT leadership of two healthcare related organisations in Hong Kong. The healthcare industry is highly sensitive about databases as they hold a patients’ personal details and medical history. 

Michael Fung, the CIO of the Chinese University of Hong Kong Medical Centre says that while the hospital mostly has proprietary software running the key hospital applications, they are now running open source databases that are typically bundled with commercial solutions, such as pharmacy applications.

  1. Benefit to the business

Notwithstanding new features, an organisations’ key considerations in adopting an open source database should depend on the benefit it will bring to the business. For organisations that want to digitally transform their business, they must also decide what outcome they want when they select their technology tools. 

“When deciding to use an open source database, the ICT teams should ensure that it is an enterprise-grade solution with a predictable life-cycle, and one where the underlying components and technologies are stable and have a known product journey,” noted Glen Francis, the Chief Technology Officer of Singapore publishing firm Singapore Press Holdings.

  1. Regular Upgrades 

Choose open source software that has regular product upgrades, security patches and key features like multi-zoning options and high availability.  

  1. One size doesn’t fit all

Determining which type of database software is right for an organisation, application, customer experience, and enterprise goals has become a complicated process. While open source database software has gained market traction, there is no “one size fits all” solution. 

Growing appetite for Open Source Database solutions

Frank Courtney-Jay, EDB’s Regional Sales Director stressed that more enterprises are adopting open source databases today. “Our customers have expressed that their appetite for open source is growing. Open source relieves them of the restrictions of proprietary arrangements and offers a more economical, agile, and flexible alternative that enables faster innovation.”

Pointing to research from a 2019 Open Source Database Report, he said PostgreSQL is extremely popular among enterprises. “One of the key benefits of using PostgreSQL is that it offers freedom to its users as there is no platform or vendor-lock in. You can run PostgreSQL anywhere, whether it is a data centre, on the cloud, or in containers, and you can get support from a vendor or no vendor.”

Ultimately, among enterprises the reality is that while open source database software is being embraced, the database landscape is still a blend of traditional and open source solutions. Organisations need to match the solution to expected outcomes in their digital transformation journey.

Mark Ko

Mark Ko

Besides tech, I love chicken rice. Point me in the right direction and I'll go and try it. :)
ViewQwest is Singapore’s fastest Internet service provider for three years in a row Previous post ViewQwest is Singapore’s fastest Internet service provider for three years in a row
Nokia, Elisa and Qualcomm achieve 5G speed record in Finland Next post Nokia to lead the EU’s 6G project Hexa-X

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: