Safely Modernize Legacy Systems with Palantir Foundry Container Engine (FCE)

By Madeline Zimmerman, R&D Federal Deployment Strategist, Palantir


Missile warnings. Airplane flight statuses. Satellite observation alerts. Much of the U.S. Government’s most critical digital infrastructure is dependent on software built during the Cold War, written in archaic languages (e.g., Fortran, COBOL, ADA), and/or installed exclusively on mainframe computers. While the infrastructure is old and may struggle to keep up with the needs of today, the core logic often works well. Yet re-writing decades of work and millions of lines of code to try to modernize just isn’t feasible.

Introducing Foundry Container Engine (FCE): FCE runs containerized legacy code in Foundry, enabling government agencies to leverage what’s working and safely leave behind what isn’t. For an analogy, consider AWS Lambda, which revolutionized how engineers run code by abstracting away the hardware infrastructure required — no more worrying about servers and clusters. In a similar fashion, FCE is revolutionizing how engineers integrate and orchestrate legacy investments in their modern software architectures. FCE streamlines your modernization journey, allowing you to incrementally rebuild millions of lines of legacy code while continuously delivering new value to the organization.


The Challenge: Operationalizing legacy code is hard

Code that is decades old is not inherently bad. On the contrary, it’s been battle-tested over decades and written by people with deep expertise in highly specialized fields. Yet aligning old software to the changing operational realities of today is both daunting and necessary. It’s often untenable work to re-write and scale up the satellite model that was built to detect 100 satellites in the 1980s and now needs to detect 30,000 satellites in 2023.

Our customers who rely on legacy code and infrastructure frequently face the following challenges:

  • Modernizing is disruptive: Too often, the only options presented for modernization are highly disruptive — data lakes, code base overhauls, and multi-year roadmaps. These run the risk of taking critical systems offline without ever accomplishing the necessary operational outcomes.

  • Unscalable: There is a long lead time for up-sizing environments to meet computation requirements, and scaling replicas of instances of products is often impossible. Forget about doing this in real-time to meet today’s critical deadlines.

  • Siloed logic: Sophisticated legacy models are much more valuable when integrated with other data sources. In our satellite example, this might include observation data, surveillance networks, sensors, and more. Adding new data feeds, data processes, outputs, and interfaces is unfeasible or too slow to be valuable.

  • Closed ecosystem: The data pipelines associated with legacy code are often a black box. There is no way for other platform and development teams to securely collaborate, effectively limiting upside by restricting the number of people able to interact with the code and provide novel analyses.

  • Divorced from operational decisions: Legacy models produce compelling insights, but the outputs are not actionable. There is no easy way to automatically create an intuitive visualization or useful alerting logic. A collision model might show satellites are about to collide, but users cannot action this information to re-orient where those satellites are flying.

Solution: Use FCE to lift and shift logic to Foundry and make it 100x more valuable

As a centrally-managed, cloud-based SaaS platform, Foundry offers instant access to cutting edge modern software implementation, including streaming pipelines, live API-driven inference, and autoscaling. Now that FCE allows containerized code to run in Foundry, unlocking the full value of legacy systems has never been easier.

Day 1 benefits include:

  • Safely and incrementally modernize: Immediately start your modernization journey with the assurance that critical systems will continue to function, and deprecation of old components will do no harm.

  • Rapidly scalable infrastructure: Achieve on-demand expansion of your compute and storage environment as capabilities evolve and expand. This provides resiliency and redundancy to avoid a single point of failure. Replacing one file with another in the FAA’s flight software should not cause flights to be grounded nationwide.

  • Flexibility and interoperability: Seamless addition of future data feeds, data processes, objects, schemas, and interfaces. Fuse disparate data to quickly produce new analyses.

  • Secure collaboration: Built-in security access control features enable secure collaboration among combined platform and development teams. When combined with pipeline transparency and DevSecOps iteration, customers can securely democratize outputs over open, extensible APIs.

  • Modern and dynamic user interfaces for rapid and automated decision-making: Users easily configure alerting logic and produce new applications with low-code/no-code tooling. Translate the complex output of a satellite physics model into an operationally relevant Space Domain Awareness application.

In a silo, legacy software can still be improved, but those small gains come at the expense of the significant, compounding benefits of modernization. FCE enables agencies to rapidly speed up progress towards their software-driven outcomes by integrating anything run by FCE with other Foundry products (e.g., pipeline builder, streaming, workshop). With Foundry’s core principles of modularity and interoperability, agencies can selectively deprecate legacy software components without disrupting their data sources, ontology, and actions. In a world where the missiles are parabolic one month and hypersonic the next, innovation in bits must outpace innovation in atoms.

This post originally appeared on Palantir.com and is re-published with permission.

Download our Resource, "Impact Study: Accelerating Interoperability with Palantir Foundry" to learn more about how Palantir Technologies can support your organization.