Facebook Folly Is Designed To Optimized Data Structures & Concurrency Primitives, but HFT Firms don't use it. WHY?

Facebook Folly is a powerful C++ library that provides several useful?data structures?and concurrency primitives. However, it may not be the best fit for all High-Frequency Trading (HFT) use cases due to several reasons:

1. Performance: HFT requires extremely?low latency?and high throughput, and any additional overhead can impact the performance. While Folly is designed to be high-performance, it may not provide the level of performance required by HFT applications.
2. Complexity: HFT applications are often designed to be simple and lightweight to reduce the risk of failures. Folly adds additional dependencies and complexity to the software stack, which can make it harder to maintain and troubleshoot.
3. Customization: HFT firms often use highly optimized and?customized software stacks?to achieve the lowest possible latency. While Folly provides several useful data structures and?concurrency primitives, it may not be the best fit for all HFT use cases, where custom data structures and algorithms are often required.
4. Reliability: HFT systems require high reliability and?fault tolerance?to ensure that trades are processed correctly and without delay. While Folly is designed to be reliable, it may not provide the level of fault tolerance required by HFT applications.

Overall, HFT firms may prefer to use in-house solutions or well-established, widely-used libraries that have a proven track record of performance and reliability in HFT environments, rather than adopting newer libraries like Folly that may not have been extensively tested in the HFT domain.

HFT firms often have strict requirements around?code quality?and maintainability. While?Facebook Folly?is a well-designed library with a lot of useful features, it may not meet the specific code quality and maintainability requirements of HFT firms.

For example, HFT firms may have specific?coding standards?and guidelines that are not compatible with Folly's coding style or design choices. As a result, HFT firms may prefer to use libraries that adhere to their specific coding standards and requirements or develop their own libraries in-house that are tailored to their specific needs.