Qilong Pan

Qilong Pan, Sameh Abdulah, Mustafa Abduljabbar et al.

Emulating computationally intensive scientific simulations is crucial for enabling uncertainty quantification, optimization, and informed decision-making at scale. Gaussian Processes (GPs) offer a flexible and data-efficient foundation for statistical emulation, but their poor scalability limits applicability to large datasets. We introduce the Scaled Block Vecchia (SBV) algorithm for distributed GPU-based systems. SBV integrates the Scaled Vecchia approach for anisotropic input scaling with the Block Vecchia (BV) method to reduce computational and memory complexity while leveraging GPU acceleration techniques for efficient linear algebra operations. To the best of our knowledge, this is the first distributed implementation of any Vecchia-based GP variant. Our implementation employs MPI for inter-node parallelism and the MAGMA library for GPU-accelerated batched matrix computations. We demonstrate the scalability and efficiency of the proposed algorithm through experiments on synthetic and real-world workloads, including a 50M point simulation from a respiratory disease model. SBV achieves near-linear scalability on up to 512 A100 and GH200 GPUs, handles 2.56B points, and reduces energy use relative to exact GP solvers, establishing SBV as a scalable and energy-efficient framework for emulating large-scale scientific models on GPU-based distributed systems.

Qilong Pan

Visually evaluating the goodness of generated Multivariate Time Series (MTS) are difficult to implement, especially in the case that the generative model is Generative Adversarial Networks (GANs). We present a general framework named Gaussian GANs to visually evaluate GANs using itself under the MTS generation task. Firstly, we attempt to find the transformation function in the multivariate Kolmogorov Smirnov (MKS) test by explicitly reconstructing the architecture of GANs. Secondly, we conduct the normality test of transformed MST where the Gaussian GANs serves as the transformation function in the MKS test. In order to simplify the normality test, an efficient visualization is proposed using the chi square distribution. In the experiment, we use the UniMiB dataset and provide empirical evidence showing that the normality test using Gaussian GANs and chi sqaure visualization is effective and credible.

Jihao Xin, Tian Lyu, Qilong Pan et al.

Post-training compression reduces LLM parameter counts but often produces irregular tensor dimensions that degrade GPU performance -- a phenomenon we call \emph{dimensional misalignment}. We present a full-stack analysis tracing root causes at three levels: framework, library, and hardware. The key insight is that model inference becomes slower because the resulting dimensions are unfriendly with the GPU execution stack. For example, compressing Llama-3-8B with activation-aware singular value decomposition (ASVD) has 15\% fewer parameters yet runs no faster than the uncompressed baseline, because 95\% of its dimensions are misaligned. We propose \textbf{GAC} (GPU-Aligned Compression), a new compression paradigm that wraps any dimension-reducing compressor and re-selects hardware-aligned dimensions via multi-choice knapsack optimization under the same parameter budget. We evaluate GAC on Llama-3-8B with ASVD and LLM-Pruner, achieving 100\% alignment and recovering up to 1.5$\times$ speedup while preserving model quality.