Hyesoon Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hyesoon Kim has a total of 85 co-authors. Here are the top co-authors by number of shared publications:
- Ramyad Hadidi
Georgia Institute of Technology, USA
7 shared publications - Jiashen Cao
Georgia Institute of Technology
6 shared publications - Bahar Asgari
University of Maryland, College Park, College Park, MD, USA
4 shared publications - Ruobing Han
Georgia Institute of Technology, Atlanta, USA
4 shared publications - Tushar Krishna
Georgia Institute of Technology, Atlanta, GA, USA
4 shared publications - Andrei Bersatti
Georgia Institute of Technology, Atlanta, GA, USA
3 shared publications
Latest Publications
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (2) , 387-390
Making Locality-Aware GEMM Compatible With Page-Granularity Placement on Chiplet GPUs
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (2) , 387-390
Making Locality-Aware GEMM Compatible With Page-Granularity Placement on Chiplet GPUs
- Open Access
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 355-368
Scaling GPU-to-CPU Migration for Efficient Distributed Execution on CPU Clusters
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 355-368
Open AccessScaling GPU-to-CPU Migration for Efficient Distributed Execution on CPU Clusters
- Open Access
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 155-167
Inside VOLT: Designing an Open-Source GPU Compiler (Tool)
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 155-167
Open AccessInside VOLT: Designing an Open-Source GPU Compiler (Tool)
- Open Access
CuFuzz: Hardening CUDA Programs through Transformation and Fuzzing
GPUs have gained significant popularity over the past decade, extending beyond their original role in graphics rendering. This evolution has brought GPU security and reliability to the forefront of concerns. Prior research has shown that CUDA's lack of memory safety can lead to serious vulnerabilities. While fuzzing is effective for finding such bugs on CPUs, equivalent tools for GPUs are lacking due to architectural differences and lack of built-in error detection. In this paper, we propose CuFuzz, a novel compiler-runtime co-design solution to extend state-of-the-art CPU fuzzing tools to GPU programs. CuFuzz transforms GPU programs into CPU programs using compiler IR-level transformations to enable effective fuzz testing. To the best of our knowledge, CuFuzz is the first mechanism to bring fuzzing support to CUDA, addressing a critical gap in GPU security research. By leveraging CPU memory error detectors such as Address Sanitizer, CuFuzz aims to uncover memory safety bugs and related correctness vulnerabilities in CUDA code, enhancing the security and reliability of GPU-accelerated applications. To ensure high fuzzing throughput, we introduce two compiler-runtime co-optimizations tailored for GPU code: Partial Representative Execution (PREX) and Access-Index Preserving Pruning (AXIPrune), achieving average throughput improvements of 32x with PREX and an additional 33% gain with AXIPrune on top of PREX-optimized code. Together, these optimizations can yield up to a 224.31x speedup. In our fuzzing campaigns, CuFuzz uncovered 122 security vulnerabilities in widely used benchmarks.
Open AccessCuFuzz: Hardening CUDA Programs through Transformation and Fuzzing
GPUs have gained significant popularity over the past decade, extending beyond their original role in graphics rendering. This evolution has brought GPU security and reliability to the forefront of concerns. Prior research has shown that CUDA's lack of memory safety can lead to serious vulnerabilities. While fuzzing is effective for finding such bugs on CPUs, equivalent tools for GPUs are lacking due to architectural differences and lack of built-in error detection. In this paper, we propose CuFuzz, a novel compiler-runtime co-design solution to extend state-of-the-art CPU fuzzing tools to GPU programs. CuFuzz transforms GPU programs into CPU programs using compiler IR-level transformations to enable effective fuzz testing. To the best of our knowledge, CuFuzz is the first mechanism to bring fuzzing support to CUDA, addressing a critical gap in GPU security research. By leveraging CPU memory error detectors such as Address Sanitizer, CuFuzz aims to uncover memory safety bugs and related correctness vulnerabilities in CUDA code, enhancing the security and reliability of GPU-accelerated applications. To ensure high fuzzing throughput, we introduce two compiler-runtime co-optimizations tailored for GPU code: Partial Representative Execution (PREX) and Access-Index Preserving Pruning (AXIPrune), achieving average throughput improvements of 32x with PREX and an additional 33% gain with AXIPrune on top of PREX-optimized code. Together, these optimizations can yield up to a 224.31x speedup. In our fuzzing campaigns, CuFuzz uncovered 122 security vulnerabilities in widely used benchmarks.
- Open Access
- 17 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1397-1411
Swift and Trustworthy Large-Scale GPU Simulation with Fine-Grained Error Modeling and Hierarchical Clustering
- 17 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1397-1411
Open AccessSwift and Trustworthy Large-Scale GPU Simulation with Fine-Grained Error Modeling and Hierarchical Clustering
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (2) , 257-260
Contention-Aware GPU Thread Block Scheduler for Efficient GPU-SSD
The growing dataset sizes in LLM have made low-cost SSDs a popular solution for extending GPU memory in mobile devices. In this paper, we introduce CA-Scheduler, a contention-aware scheduling scheme for GPU-initiated SSD access. The key insight behind CA-Scheduler is leveraging the BSP GPU programming model, which allows reordering work at the thread block level to optimize SSD throughput. By capitalizing on the predictable memory access patterns of GPU thread blocks, CA-Scheduler anticipates SSD locations to minimize contention and improve performance.
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (2) , 257-260
Contention-Aware GPU Thread Block Scheduler for Efficient GPU-SSD
The growing dataset sizes in LLM have made low-cost SSDs a popular solution for extending GPU memory in mobile devices. In this paper, we introduce CA-Scheduler, a contention-aware scheduling scheme for GPU-initiated SSD access. The key insight behind CA-Scheduler is leveraging the BSP GPU programming model, which allows reordering work at the thread block level to optimize SSD throughput. By capitalizing on the predictable memory access patterns of GPU thread blocks, CA-Scheduler anticipates SSD locations to minimize contention and improve performance.
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Buffer Management for Out-of-GPU LLM Execution
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessBuffer Management for Out-of-GPU LLM Execution
- Open Access
- 17 June 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (3) , 1-27
Aero: Adaptive Query Processing of ML Queries
Query optimization is critical in relational database management systems (DBMSs) for ensuring efficient query processing. The query optimizer relies on precise selectivity and cost estimates to generate optimal query plans for execution. However, this static query optimization approach falls short for DBMSs handling machine learning (ML) queries. ML-centric DBMSs face distinct challenges in query optimization. First, performance bottlenecks shift to user-defined functions (UDFs), often encapsulating deep learning models, making it difficult to estimate UDF statistics without profiling the query. Second, optimal query plans for ML queries are data-dependent, requiring dynamic plan adjustments during execution. To address these challenges, we introduce Aero, an ML-centric DBMS that utilizes adaptive query processing (AQP) for efficiently processing ML queries. Aero optimizes the evaluation of UDF-based query predicates by dynamically adjusting predicate evaluation order and enhancing UDF execution scalability. By integrating AQP, Aero continuously monitors UDF statistics, routes data to predicates in an optimal order, and dynamically allocates resources for evaluating predicates. Aero achieves up to 6.4x speedup compared to a state-of-the-art ML-centric DBMS across four diverse use cases, with no impact on accuracy.
- 17 June 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (3) , 1-27
Open AccessAero: Adaptive Query Processing of ML Queries
Query optimization is critical in relational database management systems (DBMSs) for ensuring efficient query processing. The query optimizer relies on precise selectivity and cost estimates to generate optimal query plans for execution. However, this static query optimization approach falls short for DBMSs handling machine learning (ML) queries. ML-centric DBMSs face distinct challenges in query optimization. First, performance bottlenecks shift to user-defined functions (UDFs), often encapsulating deep learning models, making it difficult to estimate UDF statistics without profiling the query. Second, optimal query plans for ML queries are data-dependent, requiring dynamic plan adjustments during execution. To address these challenges, we introduce Aero, an ML-centric DBMS that utilizes adaptive query processing (AQP) for efficiently processing ML queries. Aero optimizes the evaluation of UDF-based query predicates by dynamically adjusting predicate evaluation order and enhancing UDF execution scalability. By integrating AQP, Aero continuously monitors UDF statistics, routes data to predicates in an optimal order, and dynamically allocates resources for evaluating predicates. Aero achieves up to 6.4x speedup compared to a state-of-the-art ML-centric DBMS across four diverse use cases, with no impact on accuracy.
- Open Access
Towards Performance-Aware Allocation for Accelerated Machine Learning on GPU-SSD Systems
The exponential growth of data-intensive machine learning workloads has exposed significant limitations in conventional GPU-accelerated systems, especially when processing datasets exceeding GPU DRAM capacity. We propose MQMS, an augmented in-storage GPU architecture and simulator that is aware of internal SSD states and operations, enabling intelligent scheduling and address allocation to overcome performance bottlenecks caused by CPU-mediated data access patterns. MQMS introduces dynamic address allocation to maximize internal parallelism and fine-grained address mapping to efficiently handle small I/O requests without incurring read-modify-write overheads. Through extensive evaluations on workloads ranging from large language model inference to classical machine learning algorithms, MQMS demonstrates orders-of-magnitude improvements in I/O request throughput, device response time, and simulation end time compared to existing simulators.
Open AccessTowards Performance-Aware Allocation for Accelerated Machine Learning on GPU-SSD Systems
The exponential growth of data-intensive machine learning workloads has exposed significant limitations in conventional GPU-accelerated systems, especially when processing datasets exceeding GPU DRAM capacity. We propose MQMS, an augmented in-storage GPU architecture and simulator that is aware of internal SSD states and operations, enabling intelligent scheduling and address allocation to overcome performance bottlenecks caused by CPU-mediated data access patterns. MQMS introduces dynamic address allocation to maximize internal parallelism and fine-grained address mapping to efficiently handle small I/O requests without incurring read-modify-write overheads. Through extensive evaluations on workloads ranging from large language model inference to classical machine learning algorithms, MQMS demonstrates orders-of-magnitude improvements in I/O request throughput, device response time, and simulation end time compared to existing simulators.
- Open Access
- 21 June 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 29 (4) , 1-25
CuPBoP: Making CUDA a Portable Language
CUDA is designed specifically for NVIDIA GPUs and is not compatible with non-NVIDIA devices. Enabling CUDA execution on alternative backends could greatly benefit the hardware community by fostering a more diverse software ecosystem. To address the need for portability, our objective is to develop a framework that meets key requirements, such as extensive coverage, comprehensive end-to-end support, superior performance, and hardware scalability. Existing solutions that translate CUDA source code into other high-level languages, however, fall short of these goals. In contrast to these source-to-source approaches, we present a novel framework, CuPBoP , which treats CUDA as a portable language in its own right. Compared to two commercial source-to-source solutions, CuPBoP offers a broader coverage and superior performance for the CUDA-to-CPU migration. Additionally, we evaluate the performance of CuPBoP against manually optimized CPU programs, highlighting the differences between CPU programs derived from CUDA and those that are manually optimized. Furthermore, we demonstrate the hardware scalability of CuPBoP by showcasing its successful migration of CUDA to AMD GPUs. To promote further research in this field, we have released CuPBoP as an open-source resource.
- 21 June 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 29 (4) , 1-25
Open AccessCuPBoP: Making CUDA a Portable Language
CUDA is designed specifically for NVIDIA GPUs and is not compatible with non-NVIDIA devices. Enabling CUDA execution on alternative backends could greatly benefit the hardware community by fostering a more diverse software ecosystem. To address the need for portability, our objective is to develop a framework that meets key requirements, such as extensive coverage, comprehensive end-to-end support, superior performance, and hardware scalability. Existing solutions that translate CUDA source code into other high-level languages, however, fall short of these goals. In contrast to these source-to-source approaches, we present a novel framework, CuPBoP , which treats CUDA as a portable language in its own right. Compared to two commercial source-to-source solutions, CuPBoP offers a broader coverage and superior performance for the CUDA-to-CPU migration. Additionally, we evaluate the performance of CuPBoP against manually optimized CPU programs, highlighting the differences between CPU programs derived from CUDA and those that are manually optimized. Furthermore, we demonstrate the hardware scalability of CuPBoP by showcasing its successful migration of CUDA to AMD GPUs. To promote further research in this field, we have released CuPBoP as an open-source resource.