← Retour au blog
tech 8 July 2026

L: A New Runtime for k and q

L redefines the efficiency of K and Q languages with a modern runtime, offering enhanced compatibility and performance.

Article inspired by the original source
l: A new runtime for k and q ↗ lv1.sh

Introduction

In the world of financial technology, the k and q programming languages play a crucial role, especially on Wall Street where their ability to process massive data volumes in record time is essential. With the arrival of the "l" runtime, these languages reach new heights of efficiency and performance. But what makes "l" so special?

Why a New Runtime?

The K and Q languages are renowned for their elegant syntax and efficient array manipulation capabilities. However, current hardware limitations, particularly bandwidth, can hinder performance. "l" introduces an innovative approach by compressing vectors and optimizing execution to fully utilize the SIMD and multi-threading capabilities of modern processors.

Full Compatibility

One of the key strengths of "l" is its full compatibility with existing k4, q, and qSQL codes. There's no need to rewrite anything: tables, dictionaries, partitions, and other structures are natively supported, ensuring a seamless transition to this new runtime.

Compressed Vectors and Direct Execution

"l" employs end-to-end vector compression, whether at rest, in memory, or in transit. This means primitives can execute directly on these compressed vectors without requiring prior decompression, optimizing cache usage and reducing the volume of data moved.

Transparent SIMD and Parallelism

One of the most impressive aspects of "l" is its ability to automatically choose the best execution path. Whether scalar, SIMD, multi-threaded, or offloaded to a GPU, "l" adapts without requiring additional annotations in the code. For instance, a simple sum calculation can be distributed across multiple threads or SIMD units to maximize processing efficiency.

Proven Efficiency

Let's take a concrete example: an 86.5 million row analytical table can be processed on a single CPU without data decompression, while maintaining the legendary speed of qSQL operations. This demonstrates the power of "l" to handle massive datasets with unparalleled efficiency.

Conclusion

In summary, "l" offers a significant upgrade for k and q languages, providing a modern runtime that maximizes current hardware capabilities while preserving compatibility with existing codes. If you're looking to optimize your financial applications or large data processing, "l" is worth exploring.

Let's discuss your project in 15 minutes.

runtime k language q language SIMD data compression
Deepthix newsletter · 100% AI · every Monday 8am

An AI agent reads tech for you.

Our AI agent scans ~200 sources per week and ships the best articles to your inbox Monday 8am. Free. One click to unsubscribe.

Visit the newsletter page →

Want to automate your operations?

Let's talk about your project in 15 minutes.

Book a call