Zhiling Lan

Xin Wang, Pietro Lodi Rizzini, Sourav Medya et al.

The Dragonfly network, with its high-radix and low-diameter structure, is a leading interconnect in high-performance computing. A major challenge is workload interference on shared network links. Parallel discrete event simulation (PDES) is commonly used to analyze workload interference. However, high-fidelity PDES is computationally expensive, making it impractical for large-scale or real-time scenarios. Hybrid simulation that incorporates data-driven surrogate models offers a promising alternative, especially for forecasting application runtime, a task complicated by the dynamic behavior of network traffic. We present \ourmodel, a surrogate model that combines graph neural networks (GNNs) and large language models (LLMs) to capture both spatial and temporal patterns from port level router data. \ourmodel outperforms existing statistical and machine learning baselines, enabling accurate runtime prediction and supporting efficient hybrid simulation of Dragonfly networks.

Zhong Zheng, Zhiling Lan, Xingfu Wu et al.

Performance prediction is essential for energy-efficient computing in heterogeneous computing systems that integrate CPUs and GPUs. However, traditional performance modeling methods often rely on exhaustive offline profiling, which becomes impractical due to the large setting space and the high cost of profiling large-scale applications. In this paper, we present OPEN, a framework consists of offline and online phases. The offline phase involves building a performance predictor and constructing an initial dense matrix. In the online phase, OPEN performs lightweight online profiling, and leverages the performance predictor with collaborative filtering to make performance prediction. We evaluate OPEN on multiple heterogeneous systems, including those equipped with A100 and A30 GPUs. Results show that OPEN achieves prediction accuracy up to 98.29\%. This demonstrates that OPEN effectively reduces profiling cost while maintaining high accuracy, making it practical for power-aware performance modeling in modern HPC environments. Overall, OPEN provides a lightweight solution for performance prediction under power constraints, enabling better runtime decisions in power-aware computing environments.

Vittorio Palladino, Gianluca Palermo, Michael E. Papka et al.

As large language models span dense, mixture-of-experts, and state-space architectures and are deployed on heterogeneous accelerators under increasingly diverse multimodal workloads, optimising inference energy has become as critical as optimizing latency and throughput. Existing approaches either treat latency as an energy proxy or rely on data-hungry black-box surrogates. Both fail under varying parallelism strategies: latency and energy optima diverge in over 20% of configurations we tested, and black-box surrogates require hundreds of profiling samples to generalize across model families and hardware. We present EnergyLens, which uses symbolic regression as a structure-discovery tool over profiling data to derive a single twelve-parameter closed-form energy model expressed in terms of system properties such as degree of parallelism, batch size, and sequence length. Unlike black-box surrogates, EnergyLens decouples tensor and pipeline parallelism contributions and separates prefill from decode energy, making its predictions physically interpretable and actionable. Fitted from as few as 50 profiling measurements, EnergyLens achieves 88.2% Top-1 configuration selection accuracy across many evaluation scenarios compared to 60.9% for the closest prior analytical baseline, matches the predictive accuracy of ensemble ML methods with 10x fewer profiling samples, and extrapolates reliably to unseen batch sizes and hardware platforms without structural modification, making it a practical, interpretable tool for energy-optimal LLM deployment.

Matthew T. Dearing, Yiheng Tao, Xingfu Wu et al.

This paper addresses the problem of providing a novel approach to sourcing significant training data for LLMs focused on science and engineering. In particular, a crucial challenge is sourcing parallel scientific codes in the ranges of millions to billions of codes. To tackle this problem, we propose an automated pipeline framework called LASSI, designed to translate between parallel programming languages by bootstrapping existing closed- or open-source LLMs. LASSI incorporates autonomous enhancement through self-correcting loops where errors encountered during the compilation and execution of generated code are fed back to the LLM through guided prompting for debugging and refactoring. We highlight the bi-directional translation of existing GPU benchmarks between OpenMP target offload and CUDA to validate LASSI. The results of evaluating LASSI with different application codes across four LLMs demonstrate the effectiveness of LASSI for generating executable parallel codes, with 80% of OpenMP to CUDA translations and 85% of CUDA to OpenMP translations producing the expected output. We also observe approximately 78% of OpenMP to CUDA translations and 62% of CUDA to OpenMP translations execute within 10% of or at a faster runtime than the original benchmark code in the same language.

Zhong Zheng, Michael E. Papka, Zhiling Lan

Modern multi GPU HPC systems expose substantial computational capacity, yet inefficient GPU allocation often leads to wasted energy and underutilization. In practice, GPU applications exhibit heterogeneous and nonlinear scaling, making it inefficient to always use all available GPUs. We present EcoSched, an online scheduler that jointly optimizes GPU count selection and application coscheduling to improve workload level efficiency on multi GPU systems. EcoSched uses lightweight runtime profiling to estimate relative performance across GPU counts, applies a score based policy to balance energy efficiency and idle resources, and incorporates NUMA aware placement to mitigate interference. We implement EcoSched on heterogeneous CPU GPU platforms and evaluate it with diverse workloads on H100, A100, and V100 systems. EcoSched achieves up to 14.8% energy savings, 30.1% makespan improvement, and 40.4% EDP reduction over baseline schedulers, with modest performance overhead. These results show that jointly selecting GPU counts and coscheduling actions is essential for efficient multi GPU workload execution.

Zhong Zheng, Michael E. Papka, Zhiling Lan

Power-constrained HPC systems increasingly run heterogeneous CPU--GPU applications under strict cluster-wide power limits. Existing cluster-wide power management policies rely on fair-share or utilization heuristics and do not capture application-specific sensitivity to CPU and GPU power caps, leading to inefficient use of reclaimed power. We present EcoShift, a performance-aware cluster-wide power management framework. EcoShift combines online performance prediction with a dynamic-programming-based allocator to distribute reclaimed power across CPU--GPU applications for maximum average performance improvement. Through emulation-based evaluation on two heterogeneous Intel CPU and NVIDIA A100/H100 GPU platforms with diverse CPU--GPU workloads, EcoShift consistently outperforms state-of-the-art policies, achieving up to 6% average performance improvement while preserving the cluster-wide power constraint.

Yiheng Tao, Yihe Zhang, Matthew T. Dearing et al.

Efficient scheduling of LLM inference tasks is essential for achieving low latency and high throughput, particularly with the growing use of reasoning-capable LLMs. Traditional strategies like First-Come-First-Serve (FCFS) often suffer from Head-of-Line (HOL) blocking, where long-running tasks delay shorter ones queued behind them. In this paper, we introduce PARS, a prompt-aware LLM task scheduler that improves serving efficiency by approximating shortest-job-first (SJF) scheduling through pairwise ranking with margin ranking loss. PARS focuses on impactful scheduling decisions and is seamlessly integrated into the state-of-the-art LLM serving system vLLM. It effectively predicts response-length-based task ordering, reducing latency with minimal overhead. Extensive experiments across multiple LLMs and real-world inference datasets show that PARS significantly improves performance, including for reasoning workloads. Furthermore, our cross-model evaluations demonstrate that the design generalizes well, enabling effective scheduling even when predictors are trained on different LLMs.

Matthew T. Dearing, Yiheng Tao, Xingfu Wu et al.

While large language models (LLMs) are increasingly used for generating parallel scientific codes, most efforts emphasize functional correctness, often overlooking performance, especially energy efficiency. We propose LASSI-EE, an automated LLM-based refactoring framework that generates energy-efficient parallel codes through a multi-stage, iterative approach integrating runtime power profiling, energy-aware prompting, self-correcting feedback loops, and an LLM-as-a-Judge agent for automated screening of code solutions. We introduce energy-reduction@k, a novel metric that quantifies expected energy reduction when generating k code candidates and selecting the most energy-efficient, enabling systematic evaluation of multi-attempt generation strategies. Evaluating 20 HeCBench applications and two miniApps on NVIDIA A100 and AMD MI100 GPUs, a single run (k=1) with LASSI-EE delivers refactored parallel codes with an average 29% expected energy reduction at an 81% pass rate, representing a 2.8x improvement over vanilla LLM prompting. Multiple runs (k=3) achieve an average 48% expected energy reduction at a 97% pass rate. These results are consistent across devices, demonstrating LASSI-EE's effectiveness across diverse hardware architectures.

Boyang Li, Zhiling Lan, Michael E. Papka

In the field of high-performance computing (HPC), there has been recent exploration into the use of deep reinforcement learning for cluster scheduling (DRL scheduling), which has demonstrated promising outcomes. However, a significant challenge arises from the lack of interpretability in deep neural networks (DNN), rendering them as black-box models to system managers. This lack of model interpretability hinders the practical deployment of DRL scheduling. In this work, we present a framework called IRL (Interpretable Reinforcement Learning) to address the issue of interpretability of DRL scheduling. The core idea is to interpret DNN (i.e., the DRL policy) as a decision tree by utilizing imitation learning. Unlike DNN, decision tree models are non-parametric and easily comprehensible to humans. To extract an effective and efficient decision tree, IRL incorporates the Dataset Aggregation (DAgger) algorithm and introduces the notion of critical state to prune the derived decision tree. Through trace-based experiments, we demonstrate that IRL is capable of converting a black-box DNN policy into an interpretable rulebased decision tree while maintaining comparable scheduling performance. Additionally, IRL can contribute to the setting of rewards in DRL scheduling.

Yuping Fan, Zhiling Lan

For decades, system administrators have been striving to design and tune cluster scheduling policies to improve the performance of high performance computing (HPC) systems. However, the increasingly complex HPC systems combined with highly diverse workloads make such manual process challenging, time-consuming, and error-prone. We present a reinforcement learning based HPC scheduling framework named DRAS-CQSim to automatically learn optimal scheduling policy. DRAS-CQSim encapsulates simulation environments, agents, hyperparameter tuning options, and different reinforcement learning algorithms, which allows the system administrators to quickly obtain customized scheduling policies.

Yuping Fan, Zhiling Lan, Taylor Childers et al.

Cluster scheduler is crucial in high-performance computing (HPC). It determines when and which user jobs should be allocated to available system resources. Existing cluster scheduling heuristics are developed by human experts based on their experience with specific HPC systems and workloads. However, the increasing complexity of computing systems and the highly dynamic nature of application workloads have placed tremendous burden on manually designed and tuned scheduling heuristics. More aggressive optimization and automation are needed for cluster scheduling in HPC. In this work, we present an automated HPC scheduling agent named DRAS (Deep Reinforcement Agent for Scheduling) by leveraging deep reinforcement learning. DRAS is built on a novel, hierarchical neural network incorporating special HPC scheduling features such as resource reservation and backfilling. A unique training strategy is presented to enable DRAS to rapidly learn the target environment. Once being provided a specific scheduling objective given by system manager, DRAS automatically learns to improve its policy through interaction with the scheduling environment and dynamically adjusts its policy as workload changes. The experiments with different production workloads demonstrate that DRAS outperforms the existing heuristic and optimization approaches by up to 45%.

Xingfu Wu, Valerie Taylor, Zhiling Lan

In this paper, we use modeling and prediction tool MuMMI (Multiple Metrics Modeling Infrastructure) and ten machine learning methods to model and predict performance and power and compare their prediction error rates. We use a fault-tolerant linear algebra code and a fault-tolerant heat distribution code to conduct our modeling and prediction study on the Cray XC40 Theta and IBM BG/Q Mira at Argonne National Laboratory and the Intel Haswell cluster Shepard at Sandia National Laboratories. Our experiment results show that the prediction error rates in performance and power using MuMMI are less than 10% for most cases. Based on the models for runtime, node power, CPU power, and memory power, we identify the most significant performance counters for potential optimization efforts associated with the application characteristics and the target architectures, and we predict theoretical outcomes of the potential optimizations. When we compare the prediction accuracy using MuMMI with that using 10 machine learning methods, we observe that MuMMI not only results in more accurate prediction in both performance and power but also presents how performance counters impact the performance and power models. This provides some insights about how to fine-tune the applications and/or systems for energy efficiency.