Ivy Peng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Ivy Peng has a total of 150 co-authors. Here are the top co-authors by number of shared publications:
- Stefano Markidis
KTH Royal Institute of Technology, Stockholm, Sweden
19 shared publications - Maya Gokhale
Lawrence Livermore National Laboratory
14 shared publications - Gabin Schieffer
KTH Royal Institute of Technology Stockholm Sweden
13 shared publications - Jacob Wahlgren
KTH Royal Institute of Technology
13 shared publications Stefano Markidis
KTH Royal Institute of Technology
13 shared publicationsErwin Laure
9 shared publications
Latest Publications
- 15 August 2026
- Published by Springer Nature
- p. 47-61
Communication Offloading on SmartNIC DPUs: A Quantitative Approach
- 15 August 2026
- Published by Springer Nature
- p. 47-61
Communication Offloading on SmartNIC DPUs: A Quantitative Approach
- 15 August 2026
- Published by Springer Nature
- p. 106-120
Closer in the Gap: Towards Portable Performance on RISC-V Vector Processors
- 15 August 2026
- Published by Springer Nature
- p. 106-120
Closer in the Gap: Towards Portable Performance on RISC-V Vector Processors
- Open Access
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 117-124
Which Superconducting Qubit Model is Good Enough?
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 117-124
Open AccessWhich Superconducting Qubit Model is Good Enough?
- Open Access
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 154-159
An Energy Characterization Study of ML operators in XNNPACK on RISC-V Vector Hardware
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 154-159
Open AccessAn Energy Characterization Study of ML operators in XNNPACK on RISC-V Vector Hardware
- 27 June 2026
- Published by Springer Nature
- p. 368-376
BVH-Accelerated Ray Tracing for High-Frequency Electromagnetic Backscattering
- 27 June 2026
- Published by Springer Nature
- p. 368-376
BVH-Accelerated Ray Tracing for High-Frequency Electromagnetic Backscattering
- 23 June 2026
- Published by Springer Nature
- p. 48-63
Not Your Usual FFT: QFT$\rightarrow $FFT via Classical Quantum-Circuit Simulation- 23 June 2026
- Published by Springer Nature
- p. 48-63
Not Your Usual FFT: QFT$\rightarrow $FFT via Classical Quantum-Circuit Simulation- Open Access
Enabling AI Deep Potentials for Ab Initio-quality Molecular Dynamics Simulations in GROMACS
State-of-the-art AI deep potentials provide ab initio-quality results, but at a fraction of the computational cost of first-principles quantum mechanical calculations, such as density functional theory. In this work, we bring AI deep potentials into GROMACS, a production-level Molecular Dynamics (MD) code, by integrating with DeePMD-kit that provides domain-specific deep learning (DL) models of interatomic potential energy and force fields. In particular, we enable AI deep potentials inference across multiple DP model families and DL backends by coupling GROMACS Neural Network Potentials with the C++/CUDA backend in DeePMD-kit. We evaluate two recent large-atom-model architectures, DPA2 that is based on the attention mechanism and DPA3 that is based on GNN, in GROMACS using four ab initio-quality protein-in-water benchmarks (1YRF, 1UBQ, 3LZM, 2PTC) on NVIDIA A100 and GH200 GPUs. Our results show that DPA2 delivers up to 4.23x and 3.18x higher throughput than DPA3 on A100 and GH200 GPUs, respectively. We also provide a characterization study to further contrast DPA2 and DPA3 in throughput, memory usage, and kernel-level execution on GPUs. Our findings identify kernel-launch overhead and domain-decomposed inference as the main optimization priorities for AI deep potentials in production MD simulations.
Open AccessEnabling AI Deep Potentials for Ab Initio-quality Molecular Dynamics Simulations in GROMACS
State-of-the-art AI deep potentials provide ab initio-quality results, but at a fraction of the computational cost of first-principles quantum mechanical calculations, such as density functional theory. In this work, we bring AI deep potentials into GROMACS, a production-level Molecular Dynamics (MD) code, by integrating with DeePMD-kit that provides domain-specific deep learning (DL) models of interatomic potential energy and force fields. In particular, we enable AI deep potentials inference across multiple DP model families and DL backends by coupling GROMACS Neural Network Potentials with the C++/CUDA backend in DeePMD-kit. We evaluate two recent large-atom-model architectures, DPA2 that is based on the attention mechanism and DPA3 that is based on GNN, in GROMACS using four ab initio-quality protein-in-water benchmarks (1YRF, 1UBQ, 3LZM, 2PTC) on NVIDIA A100 and GH200 GPUs. Our results show that DPA2 delivers up to 4.23x and 3.18x higher throughput than DPA3 on A100 and GH200 GPUs, respectively. We also provide a characterization study to further contrast DPA2 and DPA3 in throughput, memory usage, and kernel-level execution on GPUs. Our findings identify kernel-launch overhead and domain-decomposed inference as the main optimization priorities for AI deep potentials in production MD simulations.
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 68-80
QPU Micro-Kernels for Stencil Computation
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 68-80
Open AccessQPU Micro-Kernels for Stencil Computation
- Open Access
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1860-1867
An HPC-Inspired Blueprint for a Technology-Agnostic Quantum Middle Layer
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1860-1867
Open AccessAn HPC-Inspired Blueprint for a Technology-Agnostic Quantum Middle Layer
- Open Access
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1676-1683
RISC-V Vectorization Coverage for HPC: A TSVC-Based Analysis
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1676-1683
Open AccessRISC-V Vectorization Coverage for HPC: A TSVC-Based Analysis