arXiv subjects

Yingrui Wang

Publications and source records attributed to Yingrui Wang.

4 recordsLinked to original sources

arXiv2025

Characterizing Mobile SoC for Accelerating Heterogeneous LLM Inference

With the rapid advancement of artificial intelligence technologies such as ChatGPT, AI agents, and video generation, contemporary mobile systems have begun integrating these AI capabilities on local devices to enhance privacy and reduce response latency. To meet the computational demands of AI tasks, current mobile SoCs are equipped with diverse AI accelerators, including GPUs and Neural Processing Units (NPUs). However, there has not been a comprehensive characterization of these heterogeneous processors, and existing designs typically only leverage a single AI accelerator for LLM inference, leading to suboptimal use of computational resources and memory bandwidth. In this paper, we first summarize key performance characteristics of heterogeneous processors, SoC memory bandwidth, etc. Drawing on these observations, we propose different heterogeneous parallel mechanisms to fully exploit both GPU and NPU computational power and memory bandwidth. We further design a fast synchronization mechanism between heterogeneous processors that leverages the unified memory architecture. By employing these techniques, we present HeteroInfer, the fastest LLM inference engine in mobile devices which supports GPU-NPU heterogeneous execution. Evaluation shows that HeteroInfer delivers a 1.34x to 6.02x end-to-end speedup over state-of-the-art GPU-only and NPU-only LLM engines, while maintaining negligible interference with other applications.

cs.DC↗

arXiv2024

Vertical-supercooling-controlled interfacial instability for a spreading liquid film

Thermal effect is essential to regulate the interfacial instabilities for diverse technology applications. Here we report the fingering instability at the propagation front for a spreading liquid film subjected to the supercooling at the vertical direction. We find the onset timescale of hydrodynamic instability is strongly correlated with that of the vertical solidification process. This correlation is further validated in a non-uniform geometry, demonstrating the capability of controlling fingering instability by structure design. We attribute the identified interfacial instability to a pronounced thermo-viscous effect, since the rapidly increased viscosity of propagation front undergoing solidification can significantly enhance the mobility contrast locally in the vicinity of the spreading front, consequently producing the instability analogous to viscous fingering. This work offers another valuable dimension by gating the vertical temperature to exploit the interfacial stabilities and steer liquid flow, consequently shedding light on the microfluidic cooling for electronics, and the advanced functional fibers and fabrics.

physics.flu-dyn↗

arXiv2023

An integrated framework for accelerating reactive flow simulation using GPU and machine learning models

Recent progress in artificial intelligence (AI) and high-performance computing (HPC) have brought potentially game-changing opportunities in accelerating reactive flow simulations. In this study, we introduce an open-source computational fluid dynamics (CFD) framework that integrates the strengths of machine learning (ML) and graphics processing unit (GPU) to demonstrate their combined capability. Within this framework, all computational operations are solely executed on GPU, including ML-accelerated chemistry integration, fully-implicit solving of PDEs, and computation of thermal and transport properties, thereby eliminating the CPU-GPU memory copy overhead. Optimisations both within the kernel functions and during the kernel launch process are conducted to enhance computational performance. Strategies such as static data reorganisation and dynamic data allocation are adopted to reduce the GPU memory footprint. The computational performance is evaluated in two turbulent flame benchmarks using quasi-DNS and LES modelling, respectively. Remarkably, while maintaining a similar level of accuracy to the conventional CPU/CVODE-based solver, the GPU/ML-accelerated approach shows an overall speedup of over two orders of magnitude for both cases. This result highlights that high-fidelity turbulent combustion simulation with finite-rate chemistry that requires normally hundreds of CPUs can now be performed on portable devices such as laptops with a medium-end GPU.

cs.CE↗

arXiv2023

Coalescence times for critical Galton-Watson processes with immigration

Let $X^I_n$ be the coalescence time of two particles picked at random from the $n$th generation of a critical Galton-Watson process with immigration, and let $A^I_n$ be the coalescence time of the whole population in the $n$th generation. In this paper, we study the limiting behaviors of $X^I_n$ and $A^I_n$ as $n\to\infty$.

math.PR↗