Resilience Redefined: Form3's multi-cloud payment strategy

At Form3 we aim to set the standards for resilient payments processing. Customers rightly demand that all infrastructure involved in the end-to-end flow of payments is as resilient as possible. For that reason, we opt to provide the most resilient platform in industry. 

One of the prevailing industry risks in this new era of Public Cloud adoption is Cloud concentration. An increasing number of industry stakeholders are moving their systems and workloads from many separate hosting providers and data centres into a handful of Cloud Service Providers (CSPs). While they have good resiliency track records, the concentration of services from multiple stakeholders into a handful of CSPs amplifies the industry impact should there be a CSP outage. 

Another risk to industry players is CSP lock-in. This occurs because the use of proprietary cloud provider SaaS offerings means that moving services to an alternate CSP is very complicated and costly, requiring the re-engineering of components to enable services to run on an alternate CSP. To prevent lock-in, a strategy often followed is to build services in a ‘Cloud Agnostic’ manner where they do not rely on proprietary cloud SaaS offerings. This somewhat reduces the effort and complexity of moving to an alternate CSP.  

The classic approach to providing resilience is to make use of the separation of infrastructure provided by CSPs. CSPs provide very good resilience, offering multiple Availability Zones and Regions. Typically, organisations deploy a backup service in an alternate Region or, if they are in the fortunate position of operating a ‘Cloud Agnostic’ architecture, in an alternate CSP entirely. The challenge with this approach is that the failover to a system running on an alternate Region within a CSP, or an alternate CSP entirely, is generally disruptive. It requires co-ordinated cutover with all customers and even then, the there is still the significant challenge of ensuring data is replicated to the backup platform. Additionally, once the primary platform is restored, there would need to be a co-ordinated cut-back with customers and further data replication. 

To protect ourselves and our customers from the risks highlighted above, Form3 has taken a unique and industry first approach to resilience. We have fully re-designed our platform to operate across 9 availability zones (AZs) and 3 CSPs simultaneously (3 AZs in each CSP). Our new multi-cloud platform provides real time data consistency across the CSPs. That means that our customers will be able to interact with our services in multiple CSPs in an active-active configuration and will be resilient to the loss of a CSP region or even a full CSP outage. In such a situation payments processing can merely continue in the remaining CSP. As a result, there is no need for complicated cutovers, cut-backs or data replication. Since the components in each cloud are fully exercised, and data is kept consistent across CSPs during normal operation, we have confidence that the system is capable of handling disaster conditions. To achieve our multi-cloud architecture, we had to make all our services run on Cloud Agnostic technology, which reduces the reliance on any specific cloud providers. Our platform architecture ensures that Form3 do not contribute to Cloud concentration and provides a path for our customers to improve their cloud estate resilience and risk profile.