blog-banner

Finserv Cloud Migration: CEO Convo with Citi, Google Cloud

Last edited on February 26, 2025

0 minute read

    Financial services have their sights set on a resilient cloud migration strategy: According to recent research, 53% of banking and financial services companies report experiencing service disruptions at least weekly.  

    Following the widespread service outages at major banks over the last several weeks, there is increasing urgency to deploy resilient systems industry-wide. As banks and financial services providers scale up their hybrid and multicloud deployments, they often face a daunting learning curve to ensure that their infrastructure is reliable, scalable, secure, and compliant.  

    To help with this tricky transition, Bank Automation News recently hosted an info-packed webinar, “Best practices in cloud migration for financial services.” The panel featured Spencer Kimball, Cockroach Labs Co-Founder and CEO; Dmitri Furman, Citi Head of Public Cloud; and Zac Maufer, Google Cloud Managing Director of Global Regulated Industries. Vaidik Trivedi, Bank Automation News, Associate Editor was the moderator. 

    Together, the panel explored several topics essential to cloud migration for the financial services sector, including:

    • Why financial institutions are increasingly prioritizing cloud migrations

    • How to ensure consistency across multi-region and multi-cloud deployments

    • How to support data safety in multi-cloud environments while navigating regulatory frameworks

    • How database providers like Cockroach Labs facilitate AI innovation in public clouds, for key finserv use cases such as fraud detection

    • The role of the database in managing data throughout the cloud migration process

    Cloud VelocityCopy Icon

    A key conversation centered on the benefits that can be reaped by financial institutions when they move to the cloud. Following a statement by Citi’s Furman that “cloud opens up many, many opportunities for us to service our customers better,” Kimball expanded on that point. 

    “We have seen, just practically speaking, the difference in velocity when a large financial institution is able to move some of their operations into the cloud,” said Kimball. “The cloud vendors have these incredible economies of scale. Relying on that can let a bank or financial institution of any sort leapfrog the slow and steady and expensive iterative process of upgrading their own on-premise infrastructure.” 

    The benefits of a multi-cloud strategy for banks was a central topic. After Furman discussed Citi’s use of multiple cloud providers, Kimball expanded on Cockroach Labs’ role in supporting multi-cloud operations. “With the distributed architecture, CockroachDB is always seeking to maintain consistency across different replication sites. Often people run CockroachDB for a single use case in one cloud, but they might use multiple availability zones – that's probably the most common way to use it or multiple regions within that cloud.”  

    “There's also some customers that are looking to run CockroachDB just the exact same way in terms of multiple replication sites, but instead of being within a single cloud, they run one application across multiple clouds. There's an example of this, it's a payments company called Form3, they run across all three of the hyperscalers – GCP, AWS, and Azure – and they offer their services to big financial institutions to have a payment solution. That's interesting because it does allow a cloud to be taken offline and still have complete business continuity.” 

    With DORA regulations now in effect throughout Europe, regulatory concerns are very much top-of-mind for financial institutions. Trivedi asked Kimball how CockroachDB helps banks to navigate data safety issues across multiple clouds, while simultaneously supporting compliance and high availability of data.  

    After acknowledging the security excellence provided by all the cloud vendors, Kimball said, “There's a lot we can do at the database level. Product security is extremely important: There's encryption at rest, encryption in flight, key rotations like customer-managed encryption keys, row-level security, there's partitioning of all different sorts, and plenty of things beyond that like role-based access controls.”  

    “I think over a hundred countries have their own data sovereignty requirements,” he added. “So being able to control how the data is domiciled is very important. CockroachDB is a geo-aware relational database, so at a row level you're able to designate the location of a particular piece of data, and then you may decide to limit what can be accessed based on different policies and you can ensure the data's domiciled in a particular location.” 

    Maximizing AICopy Icon

    All of the panelists weighed in on the enticing possibilities of AI, with both Furman and Maufe agreeing that data infrastructure modernization is foundational to making maximum use of the technology. 

    “If you want the very latest consistent data that you're going to feed into a generative AI prompt, or into a machine learning model that, for example, does fraud detection, you do want the availability and scalability of your operational systems of record to be able to participate,” Kimball noted. “CockroachDB is also integrating a very, very scalable vector indexing capability so that you can search, for example, over billions of vectors and have a guarantee on freshness and accuracy that doesn't degrade as the data naturally is mutating in the index because the system of record is being updated continuously as, for example, people make deposits and withdrawals, and there's payment transfers and so forth. You want all that to be captured, especially if you're doing something like fraud detection, which really creates quite an attack surface if it starts to get stale in terms of what it’s tracking.”  

    Spencer Kimball Co-Founder and CEO of Cockroach Labs

    Spencer Kimball is Co-Founder and CEO of Cockroach Labs

    As a cloud-native, distributed SQL database, CockroachDB is an optimal solution for finserv organizations as they accelerate their cloud deployments and strive to modernize their data infrastructure. 

    “We’ve found that many of the enterprise companies out there – certainly in financial services – already have a multicloud footprint, so being able to span clouds or migrate easily between them is quite important,” Kimball observed in the webinar. “That’s where a distributed architecture really does pay off: You’re able to straddle these different environments with migrations that actually have no down time.”   

    “You might think about the many pieces of technology that are required for any kind of use case,” continued Kimball. “The database is a particularly important one. There’s a lot of gravity in data so if you move a database first that actually can work in these different environments, that gives you a lot of flexibility in terms of how you can carry out your migration both safely and with minimal impact to your end customers.”

    Finserv organizations can confidently plan their path to database modernization with The Architect's Guide to SQL Database Modernization: Your Step-by-Step Roadmap.