Optimizing FDTD Solvers for Electromagnetics: A Compiler-Guided Approach with High-Level Tensor Abstractions
This work addresses performance bottlenecks and portability issues in computational electromagnetics and photonics, representing an incremental improvement through compiler-guided optimization.
The paper tackled the computational challenges and lack of portability in Finite Difference Time Domain (FDTD) solvers by introducing a domain-specific compiler based on MLIR/LLVM, achieving up to 10x speedup over baseline Python implementations on various hardware platforms.
The Finite Difference Time Domain (FDTD) method is a widely used numerical technique for solving Maxwell's equations, particularly in computational electromagnetics and photonics. It enables accurate modeling of wave propagation in complex media and structures but comes with significant computational challenges. Traditional FDTD implementations rely on handwritten, platform-specific code that optimizes certain kernels while underperforming in others. The lack of portability increases development overhead and creates performance bottlenecks, limiting scalability across modern hardware architectures. To address these challenges, we introduce an end-to-end domain-specific compiler based on the MLIR/LLVM infrastructure for FDTD simulations. Our approach generates efficient and portable code optimized for diverse hardware platforms.We implement the three-dimensional FDTD kernel as operations on a 3D tensor abstraction with explicit computational semantics. High-level optimizations such as loop tiling, fusion, and vectorization are automatically applied by the compiler. We evaluate our customized code generation pipeline on Intel, AMD, and ARM platforms, achieving up to $10\times$ speedup over baseline Python implementation using NumPy.