Adapting AlphaEvolve to Optimize Fully Homomorphic Encryption on TPUs

TL;DR

This paper presents AlphaEvolve, an automated system that uses evolutionary search and LLM-driven code generation to optimize Fully Homomorphic Encryption primitives on TPUs, significantly improving performance.

Contribution

It introduces a hardware-aware optimization framework for cryptographic kernels on TPUs, combining evolutionary algorithms and LLMs to automate complex low-level hardware tuning.

Findings

AlphaEvolve improved TFHE bootstrap latency by 2.5x.

CKKS rotation latency was reduced by 1.31x.

CKKS multiplication latency was reduced by 1.18x.

Abstract

The deployment of Fully Homomorphic Encryption (FHE) at scale is hindered due to its heavy computational overhead. While specialized hardware accelerators like Google Tensor Processing Units (TPUs) can help, mapping complex cryptographic kernels onto such architectures remains a challenge. Efficient execution requires co-optimization between the systolic array-based Matrix Multiplication Unit (MXU) and Vector Processing Units (VPUs), as well as the orchestration of data movement across the vector register files. Existing compiler stacks often abstract low-level hardware utilization, requiring developers to adopt a manual trial-and-error process that often results in fragmented execution and underutilized resources. To accelerate this development process, we use AlphaEvolve to automate the exploration of hardware-aware cryptographic-kernel optimizations. We frame optimization as an evolutionary search problem, utilizing the closed-loop system provided by AlphaEvolve, that leverages LLM-driven code generation. We use real-world feedback from hardware execution and rigorous correctness testing to guide the evolution process. We evaluate AlphaEvolve optimization on primitives for both the TFHE (Jaxite) and CKKS (CROSS) FHE schemes on Google Cloud TPUv5e, a contemporary TPU architecture. Within 24 hours of automated exploration, AlphaEvolve discovered implementation-level optimizations that improve TFHE bootstrap latency by 2.5x and CKKS rotation and multiplication latency by 1.31x and 1.18x, respectively, relative to human-engineered state of the art. These results demonstrate that AlphaEvolve can be used to enable researchers to navigate the optimization trade-offs between cryptography, compilers, and hardware accelerators.