Hyojin Sung
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
Hyojin Sung has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Yongwon Shin
Pohang University of Science and Technology
4 shared publications - Eunhyeok Park
Pohang University of Science and Technology (POSTECH) Republic of Korea
3 shared publications - Jeongmin Hong
Department of Computer Science and Engineering, Pohang University of Science and Technology, Pohang, South Korea
3 shared publications - Juseong Park
POSTECH, South Korea, Pohang, Republic of Korea
3 shared publications - Euicheol Lim
SK Hynix Inc
2 shared publications - Gwangsun Kim
POSTECH, Pohang, Republic of Korea
2 shared publications
Latest Publications
- Open Access
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 893-905
Aurora: A Disaggregated GPU-PNM-PIM System for High-Throughput Mixed-Length LLM Inference
- 5 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 893-905
Open AccessAurora: A Disaggregated GPU-PNM-PIM System for High-Throughput Mixed-Length LLM Inference
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (2) , 303-306
PLUG: P rocessing-in-Memory LU T Gathering for G raph ANNS
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (2) , 303-306
PLUG: P rocessing-in-Memory LU T Gathering for G raph ANNS
- Open Access
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-14
RAPO: Retrieval-Augmented Phase Ordering
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessRAPO: Retrieval-Augmented Phase Ordering
- Open Access
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-14
HYPERF: End-to-End Autotuning Framework for High-Performance Computing
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessHYPERF: End-to-End Autotuning Framework for High-Performance Computing
- Open Access
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 899-915
ATiM: Autotuning Tensor Programs for Processing-in-DRAM
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 899-915
Open AccessATiM: Autotuning Tensor Programs for Processing-in-DRAM
- Open Access
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 168-182
NavCim: Comprehensive Design Space Exploration for Analog Computing-in-Memory Architectures
Analog computing-in-memory (ACiM) technology has shown strong potential for neural network accelerators, addressing von-Neumann performance bottlenecks with in-memory data processing and computation. Understanding the ACiM design space, including its trade-offs and constraints, and systematically and effectively exploring it for optimal performance is essential to turn the promise into a viable product. Recent research demonstrated that multi-objective searches for ACiM architectures with heterogeneous tiles can simultaneously optimize power, performance, and area (PPA), outperforming existing tiled ACiM proposals. In this paper, we propose NavCim, a comprehensive ACiM design space exploration mechanism that advances the prior work in terms of search efficiency, search space coverage, and optimization metrics. NavCim introduces predictive modeling of ACiM hardware performance and uses the PPA prediction models instead of running simulators, significantly reducing search overheads. Faster searches enable NavCim to extend the architecture and model search spaces with an evolutionary search process to optimize architectures with more than two different tile sizes for multiple input models. With accuracy-aware searches, NavCim considers PPA and model accuracy together as optimization goals to achieve more balanced trade-offs. The experimental searches show that NavCim leverages predictive models to reduce search time by up to 7.3x without compromising the quality of search results. It also successfully identifies heterogeneous ACiM architectures that can efficiently execute multiple models on a single chip, improving accuracy by up to 19% over the prior work.
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 168-182
Open AccessNavCim: Comprehensive Design Space Exploration for Analog Computing-in-Memory Architectures
Analog computing-in-memory (ACiM) technology has shown strong potential for neural network accelerators, addressing von-Neumann performance bottlenecks with in-memory data processing and computation. Understanding the ACiM design space, including its trade-offs and constraints, and systematically and effectively exploring it for optimal performance is essential to turn the promise into a viable product. Recent research demonstrated that multi-objective searches for ACiM architectures with heterogeneous tiles can simultaneously optimize power, performance, and area (PPA), outperforming existing tiled ACiM proposals. In this paper, we propose NavCim, a comprehensive ACiM design space exploration mechanism that advances the prior work in terms of search efficiency, search space coverage, and optimization metrics. NavCim introduces predictive modeling of ACiM hardware performance and uses the PPA prediction models instead of running simulators, significantly reducing search overheads. Faster searches enable NavCim to extend the architecture and model search spaces with an evolutionary search process to optimize architectures with more than two different tile sizes for multiple input models. With accuracy-aware searches, NavCim considers PPA and model accuracy together as optimization goals to achieve more balanced trade-offs. The experimental searches show that NavCim leverages predictive models to reduce search time by up to 7.3x without compromising the quality of search results. It also successfully identifies heterogeneous ACiM architectures that can efficiently execute multiple models on a single chip, improving accuracy by up to 19% over the prior work.
- Open Access
- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 142651-142667
Non-Invasive, Memory Access-Triggered Near-Data Processing for DNN Training Acceleration on GPUs
To address the off-chip bandwidth (BW) and memory capacity challenges of GPUs in DNN training, we propose a memory access-triggered near-data processing mat NDP architecture that offloads memory- and communication-bound operations. With mat NDP, normal memory accesses also serve as implicit NDP requests to enable NDP in a non-invasive manner without modifying core-side ISA/microarchitecture/SW, for practical NDP. In addition, mat NDP enables on-the-fly NDP where the data already supplied in normal memory requests for compute-bound operations are also simultaneously used for NDP; thus, matNDP can overlap even dependent kernels while also reducing memory traffic. Moreover, with the overlap, memory bandwidth (BW) underutilized by GPU cores can be used by NDP units to improve performance even under the same total memory BW. The matNDP units can be deployed to heterogeneous memory devices in a system. First, we deploy them near GPU’s memory controllers. Secondly, our NDP units are deployed in memory expanders connected to multiple GPUs to create an NDP-enabled memory eXpander Network (NDPXNet) . It can entirely offload gradient reduction and optimizer in data-parallel training, achieving additional speedups while eliminating redundancy in memory usage and optimizer execution. As a result, we 1) enable NDP without core HW/SW changes, 2) overlap the execution of dependent layers, and 3) offload both memory- and communicationbound operations from GPUs in DNN training. Through our deep learning compiler support, NDP kernels can be generated automatically without any model code modification. Consequently, mat NDP can improve training throughput by up to 2.73× and reduce energy by up to 41.4%.
- 23 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 142651-142667
Open AccessNon-Invasive, Memory Access-Triggered Near-Data Processing for DNN Training Acceleration on GPUs
To address the off-chip bandwidth (BW) and memory capacity challenges of GPUs in DNN training, we propose a memory access-triggered near-data processing mat NDP architecture that offloads memory- and communication-bound operations. With mat NDP, normal memory accesses also serve as implicit NDP requests to enable NDP in a non-invasive manner without modifying core-side ISA/microarchitecture/SW, for practical NDP. In addition, mat NDP enables on-the-fly NDP where the data already supplied in normal memory requests for compute-bound operations are also simultaneously used for NDP; thus, matNDP can overlap even dependent kernels while also reducing memory traffic. Moreover, with the overlap, memory bandwidth (BW) underutilized by GPU cores can be used by NDP units to improve performance even under the same total memory BW. The matNDP units can be deployed to heterogeneous memory devices in a system. First, we deploy them near GPU’s memory controllers. Secondly, our NDP units are deployed in memory expanders connected to multiple GPUs to create an NDP-enabled memory eXpander Network (NDPXNet) . It can entirely offload gradient reduction and optimizer in data-parallel training, achieving additional speedups while eliminating redundancy in memory usage and optimizer execution. As a result, we 1) enable NDP without core HW/SW changes, 2) overlap the execution of dependent layers, and 3) offload both memory- and communicationbound operations from GPUs in DNN training. Through our deep learning compiler support, NDP kernels can be generated automatically without any model code modification. Consequently, mat NDP can improve training throughput by up to 2.73× and reduce energy by up to 41.4%.
- 23 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 22 (1) , 61-64
XLA-NDP: Efficient Scheduling and Code Generation for Deep Learning Model Training on Near-Data Processing Memory
Deep learning (DL) model training must address the memory bottleneck to continue scaling. Processing-in-memory approaches can be a viable solution as they move computations near or into the memory, reducing substantial data movement. However, to deploy applications on such hardware, end-to-end software support is crucial for efficient computation mapping and scheduling as well as extensible code generation, but no consideration has been made for DL training workloads. In this paper, we propose XLA-NDP, a compiler and runtime solution for NDPX, a near-data processing (NDP) architecture integrated with an existing DL training framework. XLA-NDP offloads NDPX kernels and schedules them to overlap with GPU kernels to maximize parallelism based on GPU and NDPX costs, while providing a template-based code generator with low-level optimizations. The experiments showed that XLA-NDP provides up to 1.41x speedup (1.24x on average) over the GPU baseline for four DL model training.
- 23 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 22 (1) , 61-64
XLA-NDP: Efficient Scheduling and Code Generation for Deep Learning Model Training on Near-Data Processing Memory
Deep learning (DL) model training must address the memory bottleneck to continue scaling. Processing-in-memory approaches can be a viable solution as they move computations near or into the memory, reducing substantial data movement. However, to deploy applications on such hardware, end-to-end software support is crucial for efficient computation mapping and scheduling as well as extensible code generation, but no consideration has been made for DL training workloads. In this paper, we propose XLA-NDP, a compiler and runtime solution for NDPX, a near-data processing (NDP) architecture integrated with an existing DL training framework. XLA-NDP offloads NDPX kernels and schedules them to overlap with GPU kernels to maximize parallelism based on GPU and NDPX costs, while providing a template-based code generator with low-level optimizations. The experiments showed that XLA-NDP provides up to 1.41x speedup (1.24x on average) over the GPU baseline for four DL model training.
- 13 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 21 (2) , 33-36
Runtime Support for Accelerating CNN Models on Digital DRAM Processing-in-Memory Hardware
Processing-in-memory (PIM) provides promising solutions to the main memory bottleneck by placing computational logic in or near memory devices to reduce data movement overheads. Recent work explored how commercial DRAM can feature digital PIM logic while meeting fab-level energy and area constraints, and showed a significant speedup in the inference time of data-intensive deep learning models. However, convolutional neural network (CNN) models were not considered as main targets for the commercial DRAM-PIM due to their compute-intensive convolution layers. Moreover, recent studies revealed that the area and power constraints on memory die prevent DRAM-PIM from competing with GPUs and specialized accelerators in accelerating them. Recently, mobile CNN models have increasingly adopted a composition of depthwise and pointwise convolutions instead of such compute-intensive convolutions to reduce computation cost without accuracy drop. In this paper, we show that 1x1 convolution can be offloaded for PIM acceleration with integrated runtime support and without any hardware or algorithm changes. We provide further speedup with parallel execution on GPU and DRAM-PIM and code generation optimizations. Our solution achieves up to 35.2% (31.6% on average) speedup for all 1x1 convolutions for mobile CNN models against GPU.
- 13 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 21 (2) , 33-36
Runtime Support for Accelerating CNN Models on Digital DRAM Processing-in-Memory Hardware
Processing-in-memory (PIM) provides promising solutions to the main memory bottleneck by placing computational logic in or near memory devices to reduce data movement overheads. Recent work explored how commercial DRAM can feature digital PIM logic while meeting fab-level energy and area constraints, and showed a significant speedup in the inference time of data-intensive deep learning models. However, convolutional neural network (CNN) models were not considered as main targets for the commercial DRAM-PIM due to their compute-intensive convolution layers. Moreover, recent studies revealed that the area and power constraints on memory die prevent DRAM-PIM from competing with GPUs and specialized accelerators in accelerating them. Recently, mobile CNN models have increasingly adopted a composition of depthwise and pointwise convolutions instead of such compute-intensive convolutions to reduce computation cost without accuracy drop. In this paper, we show that 1x1 convolution can be offloaded for PIM acceleration with integrated runtime support and without any hardware or algorithm changes. We provide further speedup with parallel execution on GPU and DRAM-PIM and code generation optimizations. Our solution achieves up to 35.2% (31.6% on average) speedup for all 1x1 convolutions for mobile CNN models against GPU.
- 24 March 2022
- Published by Springer Nature
- p. 33-49
Hybrid Register Allocation with Spill Cost and Pattern Guided Optimization
- 24 March 2022
- Published by Springer Nature
- p. 33-49
Hybrid Register Allocation with Spill Cost and Pattern Guided Optimization