Harshitha Menon

Daniel Nichols, Aniruddha Marathe, Harshitha Menon et al.

Parallel programs in high performance computing (HPC) continue to grow in complexity and scale in the exascale era. The diversity in hardware and parallel programming models make developing, optimizing, and maintaining parallel software even more burdensome for developers. One way to alleviate some of these burdens is with automated development and analysis tools. Such tools can perform complex and/or remedial tasks for developers that increase their productivity and decrease the chance for error. Until recently, such tools for code development and performance analysis have been limited in the complexity of tasks they can perform, especially for parallel programs. However, with recent advancements in language modeling, and the availability of large amounts of open-source code related data, these tools have started to utilize predictive language models to automate more complex tasks. In this paper, we show how large language models (LLMs) can be applied to tasks specific to high performance and scientific codes. We introduce a new dataset of HPC and scientific codes and use it to fine-tune several pre-trained models. We compare several pre-trained LLMs on HPC-related tasks and introduce a new model, HPC-Coder, fine-tuned on parallel codes. In our experiments, we show that this model can auto-complete HPC functions where generic models cannot, decorate for loops with OpenMP pragmas, and model performance changes in scientific application repositories as well as programming competition solutions.

Gregory Bolet, Giorgis Georgakoudis, Konstantinos Parasyris et al.

Modern GPU software stacks demand developers who can anticipate performance bottlenecks before ever launching a kernel; misjudging floating-point workloads upstream can derail tuning, scheduling, and even hardware procurement. Yet despite rapid progress in code generation, today's Large Language Models (LLMs) are rarely tested on this kind of forward-looking reasoning. We close that gap with gpuFLOPBench, a benchmark that asks models to "count without running" by predicting single and double-precision FLOP counts for 577 CUDA kernels drawn from HeCBench, annotated with ground-truth profiles and eight execution attributes that distinguish trivially analyzable code from kernels whose FLOPs depend on hidden compiler or runtime behavior. Evaluating current closed-source reasoning models shows clear but uneven progress: the newest LLMs achieve perfect classification on straightforward kernels but still incur multiple order-of-magnitude errors whenever implicit FLOPs arise from division, intrinsic math functions, or common subexpressions. These results surface a core limitation of existing code assistants -- the inability to internalize hardware-specific microcode effects -- and position gpuFLOPBench as a focused testbed for developing LLM tooling that can reason about performance with the same rigor as experienced GPU developers. Sources are available at our repository: https://github.com/Scientific-Computing-Lab/gpuFLOPBench

Prajwal Singhania, Siddharth Singh, Lannie Dalton Hough et al.

As large language models (LLMs) continue to grow in size, distributed inference has become increasingly important. Model-parallel strategies must now efficiently scale not only across multiple GPUs but also across multiple nodes. In this work, we present a detailed performance study of multi-node distributed inference using LLMs on GPU-based supercomputers. We conduct experiments with several state-of-the-art inference engines alongside YALIS, a research-oriented prototype engine designed for controlled experimentation. We analyze the strong-scaling behavior of different model-parallel schemes and identify key bottlenecks. Since all-reduce operations are a common performance bottleneck, we develop NVRAR, a hierarchical all-reduce algorithm based on recursive doubling with NVSHMEM. NVRAR achieves up to 1.9x-3.6x lower latency than NCCL for message sizes between 128 KB and 2 MB on HPE Slingshot and InfiniBand interconnects. Integrated into YALIS, NVRAR achieves up to a 1.72x reduction in end-to-end batch latency for the Llama 3.1 405B model in multi-node decode-heavy workloads using tensor parallelism.

Md Mahbubur Rahman, Arjun Guha, Harshitha Menon

Code LLMs often default to particular programming languages and libraries under neutral prompts. We investigate whether these preferences are encoded as approximately linear directions in activation space that can be manipulated at inference time. Using a difference-in-means method, we estimate layer-wise steering vectors for five language/library pairs and add them to model hidden states during generation. Across three open-weight code LLMs, these interventions substantially increase generation toward the target ecosystem under neutral prompts and often remain effective even when prompts explicitly request the opposite choice. Steering strength varies by model and target, with common ecosystems easier to induce than rarer alternatives, and overly strong interventions can reduce output quality. Overall, our results suggest that code-style preferences in LLMs are partly represented by compact, steerable structure in activation space.

Gautam Singh, Arjun Guha, Bhavya Kailkhura et al.

Large language models have shown remarkable ability in serial code generation, but they still struggle with parallel code for which training data is comparatively scarce. A common remedy is to use coding agents that interact with external tools, but tool calls can be costly and sometimes impractical, e.g., for partially written code. We propose Parallel-Code World Models (PCWMs), reasoning LLMs that aim to predict tool outcomes directly from parallel source code. To train PCWMs, we design a novel exploration and data generation pipeline that samples diverse parallel-coding problems and candidate implementations across multiple domains, then executes them via tools to record data races and performance profiles. From these, we synthesize hindsight reasoning traces that causally connect source code to observed tool outcomes. Fine-tuning on the resulting data yields noticeable gains, with a 7B-parameter world model improving from 64.3% to 72.8% accuracy in race-outcome prediction, while an 8B-parameter model improves in a performance profiling task from 49.3% to 58.6% accuracy. Furthermore, when open-weight models were tasked with fixing data races, world-model feedback improved their race-fixing rates relative to self-feedback by 2.7%-9.1% using our 7B-parameter world model and by 6.1%-11.1% using our 14B-parameter world model. Our results suggest that reasoning models have the potential to serve as practical substitutes for external tool calls in parallel-coding agents.

Daniel Nichols, Konstantinos Parasyris, Caetano Melone et al.

As high-performance computing and AI workloads become increasingly dependent on GPUs, maintaining high performance across rapidly evolving hardware generations has become a major challenge. Developers often spend months tuning scientific applications to fully exploit new architectures, navigating a complex optimization space that spans algorithm design, source implementation, compiler flags and pass sequences, and kernel launch parameters. Existing approaches can effectively search parts of this space in isolation, such as launch configurations or compiler settings, but optimizing across the full space still requires substantial human expertise and iterative manual effort. In this paper, we present Record-Remix-Replay (R^3), a hierarchical optimization framework that combines LLM-driven evolutionary search, Bayesian optimization, and record-replay compilation techniques to efficiently explore GPU kernel optimizations from source-level implementation choices down to compiler pass ordering and runtime configuration. By making candidate evaluation fast and scalable, our approach enables practical end-to-end search over optimization dimensions that are typically treated separately. We show that Record-Remix-Replay can optimize full scientific applications better than traditional approaches over kernel parameters and compiler flags, while also being nearly an order of magnitude faster than modern evolutionary search approaches.

Harshitha Menon, Charles F. Jekel, Kevin Korner et al.

Today's scientific challenges, from climate modeling to Inertial Confinement Fusion design to novel material design, require exploring huge design spaces. In order to enable high-impact scientific discovery, we need to scale up our ability to test hypotheses, generate results, and learn from them rapidly. We present MADA (Multi-Agent Design Assistant), a Large Language Model (LLM) powered multi-agent framework that coordinates specialized agents for complex design workflows. A Job Management Agent (JMA) launches and manages ensemble simulations on HPC systems, a Geometry Agent (GA) generates meshes, and an Inverse Design Agent (IDA) proposes new designs informed by simulation outcomes. While general purpose, we focus development and validation on Richtmyer--Meshkov Instability (RMI) suppression, a critical challenge in Inertial Confinement Fusion. We evaluate on two complementary settings: running a hydrodynamics simulations on HPC systems, and using a pre-trained machine learning surrogate for rapid design exploration. Our results demonstrate that the MADA system successfully executes iterative design refinement, automatically improving designs toward optimal RMI suppression with minimal manual intervention. Our framework reduces cumbersome manual workflow setup, and enables automated design exploration at scale. More broadly, it demonstrates a reusable pattern for coupling reasoning, simulation, specialized tools, and coordinated workflows to accelerate scientific discovery.

Daniel Nichols, Pranav Polasam, Harshitha Menon et al.

Optimizing scientific software is a difficult task because codebases are often large and complex, and performance can depend upon several factors including the algorithm, its implementation, and hardware among others. Causes of poor performance can originate from disparate sources and be difficult to diagnose. Recent years have seen a multitude of work that use large language models (LLMs) to assist in software development tasks. However, these tools are trained to model the distribution of code as text, and are not specifically designed to understand performance aspects of code. In this work, we introduce a reinforcement learning based methodology to align the outputs of code LLMs with performance. This allows us to build upon the current code modeling capabilities of LLMs and extend them to generate better performing code. We demonstrate that our fine-tuned model improves the expected speedup of generated code over base models for a set of benchmark tasks from 0.9 to 1.6 for serial code and 1.9 to 4.5 for OpenMP code.

Keita Teranishi, Harshitha Menon, William F. Godoy et al.

We discuss the challenges and propose research directions for using AI to revolutionize the development of high-performance computing (HPC) software. AI technologies, in particular large language models, have transformed every aspect of software development. For its part, HPC software is recognized as a highly specialized scientific field of its own. We discuss the challenges associated with leveraging state-of-the-art AI technologies to develop such a unique and niche class of software and outline our research directions in the two US Department of Energy--funded projects for advancing HPC Software via AI: Ellora and Durban.

Gregory Bolet, Giorgis Georgakoudis, Harshitha Menon et al.

Accurate determination of the performance of parallel GPU code typically requires execution-time profiling on target hardware -- an increasingly prohibitive step due to limited access to high-end GPUs. This paper explores whether Large Language Models (LLMs) can offer an alternative approach for GPU performance prediction without relying on hardware. We frame the problem as a roofline classification task: given the source code of a GPU kernel and the hardware specifications of a target GPU, can an LLM predict whether the GPU kernel is compute-bound or bandwidth-bound? For this study, we build a balanced dataset of 340 GPU kernels, obtained from HeCBench benchmark and written in CUDA and OpenMP, along with their ground-truth labels obtained via empirical GPU profiling. We evaluate LLMs across four scenarios: (1) with access to profiling data of the kernel source, (2) zero-shot with source code only, (3) few-shot with code and label pairs, and (4) fine-tuned on a small custom dataset. Our results show that state-of-the-art LLMs have a strong understanding of the Roofline model, achieving 100% classification accuracy when provided with explicit profiling data. We also find that reasoning-capable LLMs significantly outperform standard LLMs in zero- and few-shot settings, achieving up to 64% accuracy on GPU source codes, without profiling information. Lastly, we find that LLM fine-tuning will require much more data than what we currently have available. This work is among the first to use LLMs for source-level roofline performance prediction via classification, and illustrates their potential to guide optimization efforts when runtime profiling is infeasible. Our findings suggest that with better datasets and prompt strategies, LLMs could become practical tools for HPC performance analysis and performance portability.

Lois Curfman McInnes, Dorian Arnold, Prasanna Balaprakash et al.

This report summarizes insights from the 2025 Workshop on Next-Generation Ecosystems for Scientific Computing: Harnessing Community, Software, and AI for Cross-Disciplinary Team Science, which convened more than 40 experts from national laboratories, academia, industry, and community organizations to chart a path toward more powerful, sustainable, and collaborative scientific software ecosystems. To address urgent challenges at the intersection of high-performance computing (HPC), AI, and scientific software, participants envisioned agile, robust ecosystems built through socio-technical co-design--the intentional integration of social and technical components as interdependent parts of a unified strategy. This approach combines advances in AI, HPC, and software with new models for cross-disciplinary collaboration, training, and workforce development. Key recommendations include building modular, trustworthy AI-enabled scientific software systems; enabling scientific teams to integrate AI systems into their workflows while preserving human creativity, trust, and scientific rigor; and creating innovative training pipelines that keep pace with rapid technological change. Pilot projects were identified as near-term catalysts, with initial priorities focused on hybrid AI/HPC infrastructure, cross-disciplinary collaboration and pedagogy, responsible AI guidelines, and prototyping of public-private partnerships. This report presents a vision of next-generation ecosystems for scientific computing where AI, software, hardware, and human expertise are interwoven to drive discovery, expand access, strengthen the workforce, and accelerate scientific progress.

Meir H. Shachar, Dane M. Sterbentz, Harshitha Menon et al.

Inertial fusion energy promises nearly unlimited, clean power if it can be achieved. However, the design and engineering of fusion systems requires controlling and manipulating matter at extreme energies and timescales; the shock physics and radiation transport governing the physical behavior under these conditions are complex requiring the development, calibration, and use of predictive multiphysics codes to navigate the highly nonlinear and multi-faceted design landscape. We hypothesize that artificial intelligence reasoning models can be combined with physics codes and emulators to autonomously design fusion fuel capsules. In this article, we construct a multi-agent system where natural language is utilized to explore the complex physics regimes around fusion energy. The agentic system is capable of executing a high-order multiphysics inertial fusion computational code. We demonstrate the capacity of the multi-agent design assistant to both collaboratively and autonomously manipulate, navigate, and optimize capsule geometry while accounting for high fidelity physics that ultimately achieve simulated ignition via inverse design.

Daniel Nichols, Konstantinos Parasyris, Harshitha Menon et al.

Code LLMs have become extremely popular recently for modeling source code across a variety of tasks, such as generation, translation, and summarization. However, transformer-based models are limited in their capabilities to reason through structured, analytical properties of code, such as control and data flow. Previous work has explored the modeling of these properties with structured data and graph neural networks. However, these approaches lack the generative capabilities and scale of modern LLMs. In this work, we introduce a novel approach to combine the strengths of modeling both code as text and more structured forms.

Jan Hückelheim, Harshitha Menon, William Moses et al.

Automatic differentiation, also known as backpropagation, AD, autodiff, or algorithmic differentiation, is a popular technique for computing derivatives of computer programs accurately and efficiently. Sometimes, however, the derivatives computed by AD could be interpreted as incorrect. These pitfalls occur systematically across tools and approaches. In this paper we broadly categorize problematic usages of AD and illustrate each category with examples such as chaos, time-averaged oscillations, discretizations, fixed-point loops, lookup tables, and linear solvers. We also review debugging techniques and their effectiveness in these situations. With this article we hope to help readers avoid unexpected behavior, detect problems more easily when they occur, and have more realistic expectations from AD tools.

Harshitha Menon, Konstantinos Parasyris, Tom Scogland et al.

Modern software is incredibly complex. A typical application may comprise hundreds or thousands of reusable components. Automated package managers can help to maintain a consistent set of dependency versions, but ultimately the solvers in these systems rely on constraints generated by humans. At scale, small errors add up, and it becomes increasingly difficult to find high-fidelity configurations. We cannot test all configurations, because the space is combinatorial, so exhaustive exploration is infeasible. In this paper, we present Reliabuild, an auto-tuning framework that efficiently explores the build configuration space and learns which package versions are likely to result in a successful configuration. We implement two models in Reliabuild to rank the different configurations and use adaptive sampling to select good configurations with fewer samples. We demonstrate Reliabuild's effectiveness by evaluating 31,186 build configurations of 61 packages from the Extreme-scale Scientific Software Stack(E4S). Reliabuild selects good configurations efficiently. For example, Reliabuild selects 3X the number of good configurations in comparison to random sampling for several packages including Abyss, Bolt, libnrm, OpenMPI. Our framework is also able to select all the high-fidelity builds in half the number of samples required by random sampling for packages such as Chai, OpenMPI, py-petsc4py, and slepc. We further use the model to learn statistics about the compatibility of different packages, which will enable package solvers to better select high-fidelity build configurations automatically.