MahiMarkets Opens Up New Service and Revenue Streams with MFXPulse Using AWS

MahiMarkets provides innovative financial and investment solutions to brokers, banks, and payment providers worldwide. It previously used AWS solutions to develop two of its key products: MFXCompass, a real-time risk management and trading system, and MFXEcho, a multitenant software-as-a-service analytics solution. The most recent addition to the suite is MFXPulse, a data querying tool kit that provides trade data, pricing data, and risk metrics (such as profit and loss, volume, and value at risk) at the push of a button.

The ideation for MFXPulse began in response to an internal need for effective market data access and querying for research and development purposes. The team quickly realized that this solution could greatly benefit external clients as well. However, in order to be effective, MahiMarkets needed a solution that could flexibly scale to record millions of market data points, store clients’ historical data for regulatory purposes, and be compatible with clients’ third-party tooling. To progress with this development, MahiMarkets turned to AWS for support.

MahiMarkets was able to achieve the scalability needed for MFXPulse by migrating to a serverless solution on Amazon Web Services (AWS). “By going serverless, we have virtually no limits on scalability,” says James Furness, chief technology officer of MahiMarkets. “The combination of AWS services working in sync provides great scale and low costs.”

Using its previous solution, MFXPulse was only able to store 12 weeks of historical data, but clients expressed a need to store data for longer durations due to regulatory requirements. MahiMarkets addressed this need using Amazon Simple Storage Service (Amazon S3) - an object storage service that offers industry-leading scalability, data availability, security, and performance - providing a more cost-effective solution for storing large amounts of data. “Clients can keep 1 year–plus of data on MFXPulse at a fraction of the previous cost,” says Furness.

During the development of MFXPulse, the team refined its product plan using AWS services. Development was accomplished by a two-person team - an achievement facilitated by the ease of use of the AWS Cloud and AWS managed services. MahiMarkets found that the seamless co-integration of AWS services helped significantly reduce time to market. Support from the wider AWS team meant MahiMarket’s developers did not have to spend much time building an architecture from scratch. “Working alongside the AWS Solutions Architects, we chose the optimal technologies to develop our solution,” says Furness.

Previously, MahiMarkets had experimented with various server solutions, which did not provide a level of computing power adequate for batch processing. AWS services like AWS Lambda - a serverless, event-driven compute service—helped MahiMarkets process the large volumes of market data that flow through MFXPulse while maintaining low costs by not overprovisioning resources. The new solution only costs about 20 percent of the original solution’s cost while storing more than double the amount of data. “Traditional servers are either too small to handle the workload or too big, which leads to idling resources,” says Liam Cordelle, senior software developer at MahiMarkets. “AWS Lambda automatically provisions what is needed to process the data and then shuts down until the next batch is ready to go.”

MahiMarkets was able to develop the querying functionality of MFXPulse using AWS Glue - a service to discover, prepare, move, and integrate data from multiple sources - and Amazon Athena, a serverless, interactive analytics service. “The combination of these services helped us to transform our data and provide a serverless SQL-based querying engine to interact with the data,” says Cordelle. “Clients can use our solution in different ways, depending on their use cases.” The compatibility of MFXCompass, MFXEcho, and MFXPulse has provided clients with a more seamless and transparent experience when accessing their data. All queries from MFXEcho to MFXPulse are completed in under 3 minutes. “We have created this Rosetta stone of an open-format data lake that nearly anyone can access,” says Furness. “Our clients can have their internal teams develop analysis and automated queries on that data.”

As a result of the successful integration and adoption of MFXPulse, MahiMarkets is looking to further use AWS services to deliver even more innovative solutions to its clients, including through artificial intelligence and machine learning. MFXPulse has opened a variety of new opportunities for the business. “We’ve unlocked a new frontier by getting deeper into machine learning and artificial intelligence. We’re able to quickly prototype new machine learning models and tools using different languages and frameworks on AWS,” says Furness. “This project has unlocked a lot for us in terms of our road map for innovation in industry, development of our products, enhancement of our models, and understanding of markets.”