Icon Solutions and MongoDB Benchmark Hits 6,000 TPS for Instant Payments

Icon Solutions and MongoDB have validated that the Icon Payments Framework, running on MongoDB Atlas, can process 6,000 end-to-end payment transactions per second with sub-second latency. For fintech leaders, this benchmark provides a blueprint for handling the massive traffic peaks and strict service level agreements required by the global shift toward instant payment schemes.

What was announced

The joint benchmark tested the Icon Payments Framework (IPF) using a reference SEPA Instant outbound payment flow. Each transaction underwent 12 distinct processing steps, including duplicate checking, message and scheme-rule validation, fraud and sanctions screening, FX retrieval, funds reservation, and booking. The system achieved a mean end-to-end latency of 0.51 seconds at peak load, which is significantly faster than the 10-second window typically allowed by instant payment schemes. The testing utilized a "ladder" approach, scaling from 500 to 6,000 transactions per second to observe throughput and database load at various stages.

Key technical findings showed that the platform handled approximately 430,000 database operations per second across two MongoDB Atlas clusters. The architecture maintained near-linear scaling, with CPU and insert rates rising in proportion to the load. Resilience testing demonstrated that the system could recover from application node failures in under 90 seconds and database failovers in roughly five seconds, both with zero data loss. The infrastructure consisted of 12 application pods on Kubernetes across three availability zones, utilizing Akka for fault-tolerant processing and Kafka to stream data to the Operational Data Store. Each database shard ran on an M80-class tier with 32 vCPUs and 128 GB of RAM.

"Banks need payment infrastructure that can adapt to an instant economy. Yet sustaining thousands of payments per second while maintaining resilience, availability, scalability and ultra-low latency is not easy, particularly for institutions relying on legacy architecture or building everything themselves. This benchmark demonstrates that IPF offers an alternative approach that has been designed to meet the unprecedented demands of an always-on world, enabling banks to lead payments forward with confidence."

Toine van Beusekom, Strategy Director at Icon Solutions.

The companies involved

Icon Solutions provides the Icon Payments Framework (IPF), an internationally proven payments development framework. The company’s technology is currently utilized by major Tier 1 financial institutions, including Citi, NatWest, and BNP Paribas. IPF is built on a cloud-native, event-driven architecture designed to replace or augment legacy systems that struggle with the "always-on" requirements of modern payment rails. The framework is specifically engineered for environments where scale and continuous availability are baseline requirements.

MongoDB is the provider of MongoDB Atlas, a multi-cloud developer data platform. In this benchmark, Atlas backed both the IPF Event Journal and the Operational Data Store. Boris Bialek serves as Vice President and Global Field Chief Technology Officer at MongoDB. The collaboration highlights the integration of distributed database technology with specialized payment processing frameworks. Other technologies involved in the stack include Kubernetes for container orchestration and Kafka for event streaming. These components collectively address the challenges of migrating traffic from traditional net-settlement schemes to instant rails, where downtime is not tolerated and traffic peaks are increasingly sharp.

What FF News has reported before

FF News has previously covered the digital transformation efforts of the Tier 1 banks that utilize Icon’s framework. This includes reporting on how DBS and Citi Complete First Weekend USD Cross-Border Payment via Swift Digital Ledger, showcasing the industry's move toward 24/7 settlement. Additionally, we have tracked Citi’s expansion into tokenized assets, such as when Citi Selects Caplight to Power Independent Pricing for New Digital Depositary Receipt Product. These developments underscore a broader trend among major institutions to adopt high-performance, programmable infrastructure to support the next generation of financial services and real-time liquidity management.

What this means

This benchmark signals a turning point for the "build vs. buy" debate in bank infrastructure. As account-to-account volumes migrate to instant rails, the technical debt of legacy platforms becomes a systemic risk. The ability to handle 6,000 transactions per second with zero data loss during failure puts immense pressure on traditional vendors who rely on monolithic architectures. It also raises questions for mid-tier banks: if Tier 1 players are adopting distributed, event-driven frameworks to achieve this scale, can smaller institutions afford to continue retrofitting aging systems? The industry is moving toward a model where infrastructure is a commodity of scale, and performance is the primary differentiator.