Matryoshka: Optimization of Dynamic Diverse Quantum Chemistry Systems via Elastic Parallelism Transformation

TL;DR

Matryoshka introduces an elastic parallelism transformation technique that dynamically realigns GPU parallel patterns, significantly accelerating quantum chemistry computations with diverse and variable workloads.

Contribution

The paper presents a novel elastically-parallel technique with transformation primitives and an integrated system to efficiently execute dynamic quantum chemistry workloads on GPUs.

Findings

Achieves up to 13.86x speedup over state-of-the-art methods.

Effectively handles dynamic diversity in quantum chemistry systems.

Provides a scalable framework for GPU acceleration of scientific computing.

Abstract

AI infrastructures, predominantly GPUs, have delivered remarkable performance gains for deep learning. Conversely, scientific computing, exemplified by quantum chemistry systems, suffers from dynamic diversity, where computational patterns are more diverse and vary dynamically, posing a significant challenge to sponge acceleration off GPUs. In this paper, we propose Matryoshka, a novel elastically-parallel technique for the efficient execution of quantum chemistry system with dynamic diversity on GPU. Matryoshka capitalizes on Elastic Parallelism Transformation, a property prevalent in scientific systems yet underexplored for dynamic diversity, to elastically realign parallel patterns with GPU architecture. Structured around three transformation primitives (Permutation, Deconstruction, and Combination), Matryoshka encompasses three core components. The Block Constructor serves as the central orchestrator, which reformulates data structures accommodating dynamic inputs and constructs fine-grained GPU-efficient compute blocks. Within each compute block, the Graph Compiler operates offline, generating high-performance code with clear computational path through an automated compilation process. The Workload Allocator dynamically schedules workloads with varying operational intensities to threads online. It achieves highly efficient parallelism for compute-intensive operations and facilitates fusion with neighboring memory-intensive operations automatically. Extensive evaluation shows that Matryoshka effectively addresses dynamic diversity, yielding acceleration improvements of up to 13.86x (average 9.41x) over prevailing state-of-the-art approaches on 13 quantum chemistry systems.