When it comes to understanding what the financial services industry is looking for with databases, there’s a shift from proprietary options to open-source software. The demand is for flexibility, scalability, no lock-ins, and the ability to function with minimal cost to quickly maximize return on investment. But how best can we support customers looking to make the transition?
To answer this question, we asked Saurabh Mathur, Senior Database Specialist at AWS, and Vikram Rai, Lead Database Migration Specialist at SourceFuse, to join our Talking Out Cloud series. Together, they shared their insights on modernization of legacy databases, addressing common challenges, and how together AWS and its Partners facilitate the best possible customer benefits.
#1 What are the key distinctions between open-source and commercial database management systems?
In layman’s terms, software that charges a license fee is commercial, while open-source software may charge for support but is community-driven. Oracle and SQL Server are examples of commercial databases, and while MySQL has enterprise versions that fall into this category, its open-source version, along with PostgreSQL and Redis, are community-driven.
Customers are seeking flexibility, scalability, and access to the best talent working with the latest technology. Cost is a significant factor, with open-source options offering savings. There’s also a trend away from vendor lock-in as customers want technology that can be used across any cloud platform, not just AWS, GCP, or Microsoft.
These factors drive the adoption of purpose-built databases rather than relying on a one-size-fits-all approach.
#2 Is database modernization a technical or a strategic move for a customer to be competitive in a data driven world?
Technologies are improving rapidly and l have seen significant enhancements in scalability, agility, and performance in recent years. Customers with outdated database technologies need to consider security, scalability, and performance as their systems approach the end of support. Additionally, not using a modern database means missing out on integrations with modern tools, which can impact efficiency.
From a strategic perspective, there are substantial cost savings. Companies are saving a lot by avoiding extra licensing costs for databases. Moving to cloud solutions like AWS also reduces the need for extra manpower to manage these databases.
Today’s world is data-driven, and having data available at the click of a button is a game-changer. This level of accessibility was unimaginable a few years ago but is now a reality. So, it’s definitely a strategic move as well.
#3 The banking industry continues to exhibit a strong preference for legacy databases - what would you attribute this to?
I think trust is a significant factor. Commercial database engines were the first to enter the industry and have been around for many years. They have a long history of providing top-notch performance, and there’s no doubt about their reliability.
The banking industry has traditionally been both buyers and builders. They have often purchased third-party ISV products or applications from vendors that are built on commercial databases, primarily Oracle and SQL Server. When a customer buys a system, such as a loan management system, it is typically an application built on top of an Oracle database or SQL Server. These solutions are then packaged and sold to large banks.
After buying several ISV products, banks often enter into large Unlimited License Agreements with Oracle. These agreements span the entire organization, and all business users and use cases start leveraging this cost. It has also been easier for organizations to find people skilled in working with Oracle databases or other commercial databases.
Trust is definitely a reason why core applications, especially in banking, remain on commercial databases. However, with AWS and databases like Aurora, we are challenging the notion that commercial databases are always the best option. We are demonstrating that there are better, easier, and cheaper options available that can provide commercial-grade performance at a fraction of the cost.
#4 What are the common challenges of database modernization? How does AWS help overcome them?
From my point of view, typical challenges can be divided into two parts: third-party application vendors and homegrown applications.
Third-party application vendors are customers who buy products from the market that are packaged with commercial-grade databases. Homegrown applications are easier and quicker to move from commercial-grade databases to open source. However, the challenge lies in the time and effort required to migrate stored procedures from SQL Server or Oracle databases.
Another challenge is gaining customer endorsement. It’s not just AWS that needs to take the lead; customers also need to be willing to embrace the change. The challenges I’ve typically come across are with packaged products. AWS is working with many of these vendors to address this. For homegrown applications and some ISV products, there might be dual ports, meaning the product can run on both commercial databases like Oracle and open-source databases like PostgreSQL or MySQL.
In these conversations, we introduce our Migration Acceleration Program (MAP) offering. This program provides funding to help offset some of the initial migration costs. We also have partners like SourceFuse who assist in the assessment and mobilization phases, helping customers to accelerate their licensing assessments with our Optimization and Licensing Assessment (OLA).
Partners, like SourceFuse, help guide customers through the MAP funding process towards open-source databases, and we have a step-by-step journey to assist customers in making the transition.
#5 What solutions does SourceFuse provide that help facilitate a smooth database modernization?
In my experience when talking to customers, the words “migration” or “modernization” sometimes scare people because of the perceived huge cost and effort involved. However, from a SourceFuse perspective, there are several points to consider.
Firstly, we work closely with AWS and, as mentioned, we secure funding for the customer, so the customer doesn’t have to pay a lot of money out of pocket. This makes it easier for customers to take the first step.
Secondly, we are Premier Partners of AWS, with dozens of migrations under our belt. This means we understand the process and can implement it with minimal disruption or impact on the customer’s workflow. We utilize tools provided by AWS, such as CD (Cloud Development) and DM (Database Migration), and we also have our own in-house tools that speed up the migration journey.
And thirdly, post-migration, we continue to work with the customer and their team to ensure they understand the changes and educate them on aspects of database management.
#6 From an AWS Partner’s perspective, what is your experience collaborating with the AWS teams?
So very recently, I had the opportunity to meet a couple of AWS Account Managers, APAC Managers, and Regional Managers, both at the AWS office and at events like database modernization MO clinics and CXO roundtables. They were extremely forthcoming in enabling us to connect with AWS customers to help them save costs and improve their application and database technologies.
It’s evident that AWS regards and recognizes their partners and is very helpful in terms of collaborating with us.
#7 How would you visualize database modernization 5 years from now?
(Saurabh) We are already seeing a lot of changes happening within the banking industry – customers are motivated to move out of commercial databases. To help prop up business on commercial databases, we’ve also been seeing a lot of partnerships with cloud providers. However, there’s one thing that always comes as a requirement by customers: cost. All customers globally are cost-sensitive and nobody wants to pay 40-50% of their IT budget on license fees; they’d rather repurpose it and do innovative things.
In five years from now, I think a few database engines that are going to pick up are PostgreSQL, which is already doing well. We’ve already seen a lot of flavors of PostgreSQL, like distributed SQL, not just from AWS but also from other competitors.
We’ll also see a lot of customers moving to open source, which is already progressing. So, five years from now, I definitely believe that about 70% of the databases might have either converted to open source from commercial databases or would be on their way to do so.
(Vikram) One problem statement, as mentioned, is obviously cost, but another one is the timeline of doing migrations. If I tell a customer it’s going to take six months to a year to modernize their database, that’s not an effective or efficient timeline. I predict that with all the AI/ML coming into the picture, there will be a push to speed up the automation process of converting databases from one to another. So, cheaper and faster.
