Many of our clients that use front-to-back trading systems are running into functionality gaps, such as an instrument they want to trade is not supported, the software does not get updated frequently enough, or the platform has scalability and performance issues. In some cases the vendor is under-investing in the system, or is attempting a lift and shift migration to the cloud that misses out on important cloud-native design advantages. All too often these shortcomings are addressed with inefficient or fragile workarounds that consume a lot of time and energy and lead to an incomplete picture of the firm’s investment positions and risks.
When these clients approach us, we lay out a considered strategy for filling functional gaps with Beacon and, if desired, gradually and gracefully migrating away from the legacy system onto our platform. This process usually involves a proof-of-concept phase, deploying Beacon to support a single instrument type or missing functionality, managing the data flow, and exploring the power of cloud computing.
What have you got to prove?
This first step is typically the hardest—deciding that a system you’ve used for years can no longer meet your needs, and breaking up is hard to do. When we engage with clients who are struggling with their legacy systems, we try to focus primarily on the immediate needs during the proof of concept stage, but still showcase our platform’s full capabilities. We realise that big-bang replacement projects are not an effective approach, so we look for a sustainable way of filling the functionality gap, not just another fragile workaround. This usually involves bi-directional data feeds between Beacon and existing systems, which is something that we excel at. We also want to start off with a quick win to reassure the decision makers that using our platform is the right answer. And most importantly, a successful implementation involves achieving 2-3 month milestones that are well documented and communicated upon completion.
Starting with a single instrument or functional gap
Once we have proven the basic principles and integration functionality, we look to address the client’s primary missing element. This could be trading an unsupported financial instrument, providing better or broader risk coverage, or using the cloud for real-time analyses. For example, one of our banking clients wants to add barrier reverse convertibles to their set of tradable instruments. Their legacy trading system could handle vanilla instruments, but not more exotic ones.
Beacon’s transparent licensed code base and customizable financial object hierarchy makes short work of defining a new financial instrument, without engineering change orders or custom development fees. We work together to determine the best implementation approach, deploy appropriate data connectors, and run defined test cases. Since Beacon uses the same data and code throughout the software development lifecycle, going from research to production does not require rewriting any models or re-testing on live data.
Dealing with data
Speaking of data, dealing with different databases and interfaces is a requirement for any effective solution. Beacon includes data connectors and tools for a wide range of sources, such as CME DataMine, with many more available to download from the Beacon App Store. For example, many of our banking clients want to integrate with existing order management databases and trading and risk management systems, so that they don’t have to replace everything all at once. Clients can populate their Beacon domain with the necessary data in a few hours or days, depending on the volume of data to be transferred. Synchronising trades with an existing deals database or order management system is facilitated by our Trade Ingestion Application, making it easy to visualise and customise the necessary field mappings.
Market data, including real-time streams, usually just means downloading a plugin from the Beacon App Store and configuring the desired data path and credentials. Comprehensive data warehouse tools enable secure interactions with most other internal and external data sources. Whichever interfaces you use, Beacon brings your data to life by connecting it with the models and analytics that give it meaning. You can now gradually extend your usage of Beacon to address other functional needs, confident that you are working with consistent and up-to-date data.
Beacon clients are using these tools and frameworks to integrate with and augment existing systems and databases—they don’t try for a complete replacement in one step. For example, using Beacon to trade new instruments that their legacy systems don’t support, reading from multiple disparate systems to build an overall risk picture for all asset classes, or building a centralised source of truth for data and code modules, to improve the consistency and reproducibility of pricing and risk calculations.
Leveraging the cloud
Large amounts of available data provide many benefits—if you can get the desired analyses. Running large sets of risk scenarios or generating complex analyses during the trading day can cause a big drain on in-house data centres, if those actions are even possible. Because most legacy vendor applications were not architected to scale horizontally, they often struggle with this aspect of shifting to the cloud, missing out on some of the key benefits, such as containerized workloads and elastic compute pools that allow you to perform massively parallel calculations. Beacon also uses an enterprise-scale dependency graph framework that tracks dependencies between data and models. When running calculations, the system identifies which inputs have changed, and only calculates values that are dependent on those inputs, making recalculations or large scenarios much more efficient.
Breaking up isn’t hard to do
Beacon was built to be in the cloud from the ground up, providing a cloud-agnostic approach that delivers full capabilities without the risks of being tied to a specific cloud provider. Our cloud management dashboards give you full control over how many resources you use and when to schedule them, enabling you to tailor compute services to your timing and budgetary requirements. With our system integration tools, you can even use Beacon as a processing engine for complex calculations and deliver results back to existing applications and even spreadsheets. When your legacy systems no longer meet your needs, breaking up with them doesn’t have to be uncomfortable.