Bernd Burgstaller
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
Bernd Burgstaller has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Yeonsoo Kim
Yonsei University, Seoul, Republic of Korea
9 shared publications - Bernhard Scholz
The University of Sydney
7 shared publications - Jeong-Geun Kim
Kyungpook National University, Daegu, Republic of Korea
6 shared publications - Jay Hwan Lee
Yonsei University, Seoul, Republic of Korea
5 shared publications - Hyunwoo Nam
Yonsei University, Seoul, Republic of Korea
4 shared publications - Johann Blieberger
Vienna University of Technology
3 shared publications
Latest Publications
- Open Access
- 20 July 2026
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 54 (1) , 120-122
LPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
In designing an effective GPU for large-scale workloads, a trustworthy and fast simulator is required to evaluate performance and explore the design space. However, existing GPU simulators suffer from long execution times due to detailed component simulation, limiting their utility for evaluating the effects of architectural modifications. It is necessary to improve the performance of a GPU simulator such that quick architecture exploration and evaluation for large-scale workloads are available, at the expense of accuracy. This paper presents LPGSim, a trace-driven and cycle-level GPU simulator. LPGSim aims to provide fast and accurate GPU simulation. To this end, LPGSim first eliminates instruction metadata that has minimal impact on simulation accuracy. Next, LPGSim parallelizes GPU simulation. LPGSim partitions a GPU architecture into three parallelizable subsystems and introduces local-clock-based parallelization. LPGSim further employs parallelization methods to achieve scalability on NUMA systems with an acceptable trade-off in accuracy. LPGSim shows 21.4%, 23.3%, and 22.7% errors across three different GPU architectures, while achieving a total 197.4x speedup over the state-of-the-art simulator.
- 20 July 2026
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 54 (1) , 120-122
Open AccessLPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
In designing an effective GPU for large-scale workloads, a trustworthy and fast simulator is required to evaluate performance and explore the design space. However, existing GPU simulators suffer from long execution times due to detailed component simulation, limiting their utility for evaluating the effects of architectural modifications. It is necessary to improve the performance of a GPU simulator such that quick architecture exploration and evaluation for large-scale workloads are available, at the expense of accuracy. This paper presents LPGSim, a trace-driven and cycle-level GPU simulator. LPGSim aims to provide fast and accurate GPU simulation. To this end, LPGSim first eliminates instruction metadata that has minimal impact on simulation accuracy. Next, LPGSim parallelizes GPU simulation. LPGSim partitions a GPU architecture into three parallelizable subsystems and introduces local-clock-based parallelization. LPGSim further employs parallelization methods to achieve scalability on NUMA systems with an acceptable trade-off in accuracy. LPGSim shows 21.4%, 23.3%, and 22.7% errors across three different GPU architectures, while achieving a total 197.4x speedup over the state-of-the-art simulator.
- Open Access
- 8 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 120-122
LPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
- 8 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 120-122
Open AccessLPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
- Open Access
- 29 May 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 10 (2) , 1-27
LPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
The demands of large-scale workloads have driven the evolution of GPUs, placing them in the mainstream scope of computing architectures. To design an effective GPU for large-scale workloads, a trustworthy simulator is required to evaluate performance and explore the design space. Additionally, GPU simulators must be fast enough to evaluate architectural modifications for large-scale workloads within a reasonable time. However, existing GPU simulators suffer from long execution times due to detailed component simulation, limiting their utility for evaluating the effects of architectural modifications on large-scale workloads. It is necessary to improve the performance of a GPU simulator such that quick architecture exploration and evaluation for large-scale workloads are available, at the expense of accuracy. This paper presents LPGSim, a trace-driven and cycle-level GPU simulator. LPGSim aims to provide fast and accurate GPU simulation. To this end, LPGSim first eliminates instruction metadata that has minimal impact on simulation accuracy. This approach enables coalescing compute instructions while preserving the dependencies to the preceding memory instruction. As a consequence, the core pipeline is simplified, improving simulation speed at a marginal cost in accuracy. Next, LPGSim parallelizes GPU simulation. LPGSim partitions a GPU architecture into three parallelizable subsystems and introduces local-clock-based parallelization to reduce the overhead of global synchronization and sequential execution paths in existing methods. LPGSim further employs parallelization methods to achieve scalability on NUMA systems with an acceptable trade-off in accuracy. Our evaluation shows a modest decrease in simulation accuracy in return for a substantial improvement in simulation speed. In terms of accuracy, LPGSim achieves 21.4%, 23.3%, and 22.7% errors across three different GPU architectures. In terms of speed, the single-threaded LPGSim simulation yields a 9.97x speedup over the state-of-the-art GPU simulator. Parallelization achieves a 19.8x speedup on a two-socket 56-thread system, for a total 197.4x speedup over the state-of-the-art simulator. Parallelization incurs an additional accuracy loss, but our experiments indicate that simulation is still reliable for large-scale workloads.
- 29 May 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 10 (2) , 1-27
Open AccessLPGSim: A Lightweight Parallel GPU Simulator Maximizing Speed with Trustworthy Simulation
The demands of large-scale workloads have driven the evolution of GPUs, placing them in the mainstream scope of computing architectures. To design an effective GPU for large-scale workloads, a trustworthy simulator is required to evaluate performance and explore the design space. Additionally, GPU simulators must be fast enough to evaluate architectural modifications for large-scale workloads within a reasonable time. However, existing GPU simulators suffer from long execution times due to detailed component simulation, limiting their utility for evaluating the effects of architectural modifications on large-scale workloads. It is necessary to improve the performance of a GPU simulator such that quick architecture exploration and evaluation for large-scale workloads are available, at the expense of accuracy. This paper presents LPGSim, a trace-driven and cycle-level GPU simulator. LPGSim aims to provide fast and accurate GPU simulation. To this end, LPGSim first eliminates instruction metadata that has minimal impact on simulation accuracy. This approach enables coalescing compute instructions while preserving the dependencies to the preceding memory instruction. As a consequence, the core pipeline is simplified, improving simulation speed at a marginal cost in accuracy. Next, LPGSim parallelizes GPU simulation. LPGSim partitions a GPU architecture into three parallelizable subsystems and introduces local-clock-based parallelization to reduce the overhead of global synchronization and sequential execution paths in existing methods. LPGSim further employs parallelization methods to achieve scalability on NUMA systems with an acceptable trade-off in accuracy. Our evaluation shows a modest decrease in simulation accuracy in return for a substantial improvement in simulation speed. In terms of accuracy, LPGSim achieves 21.4%, 23.3%, and 22.7% errors across three different GPU architectures. In terms of speed, the single-threaded LPGSim simulation yields a 9.97x speedup over the state-of-the-art GPU simulator. Parallelization achieves a 19.8x speedup on a two-socket 56-thread system, for a total 197.4x speedup over the state-of-the-art simulator. Parallelization incurs an additional accuracy loss, but our experiments indicate that simulation is still reliable for large-scale workloads.
- Open Access
- 3 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 486
Storage Replica: Accelerating the Storage Access of the Ethereum Virtual Machine
Ethereum’s smart contracts operate on directly addressable storage that is represented as tries. The performance of the Ethereum Virtual Machine (EVM) suffers from slow storage access due to trie encoding, which hampers transaction throughput and scalability. To mitigate the Ethereum storage performance bottleneck, we propose a new storage representation for the EVM that supports asynchronous trie construction. Without changing the Ethereum protocol, we add a flat representation called Storage Replica to improve performance. Storage Replica provides a fast lookup of values in the program’s main thread, while a worker thread prepares the tries for subsequent cryptographic calculations. With a storage overhead of less than 5% (i.e., 10 ), we achieve up to a 6× speedup in processing smart contracts and a 4× speedup in block commits for the initial 9 M blocks of the Ethereum blockchain.- 3 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 486
Open AccessStorage Replica: Accelerating the Storage Access of the Ethereum Virtual Machine
Ethereum’s smart contracts operate on directly addressable storage that is represented as tries. The performance of the Ethereum Virtual Machine (EVM) suffers from slow storage access due to trie encoding, which hampers transaction throughput and scalability. To mitigate the Ethereum storage performance bottleneck, we propose a new storage representation for the EVM that supports asynchronous trie construction. Without changing the Ethereum protocol, we add a flat representation called Storage Replica to improve performance. Storage Replica provides a fast lookup of values in the program’s main thread, while a worker thread prepares the tries for subsequent cryptographic calculations. With a storage overhead of less than 5% (i.e., 10 ), we achieve up to a 6× speedup in processing smart contracts and a 4× speedup in block commits for the initial 9 M blocks of the Ethereum blockchain. - Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 156846-156876
EtherVal: Smart Contract Decompiler Validation for the Ethereum Blockchain
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 156846-156876
Open AccessEtherVal: Smart Contract Decompiler Validation for the Ethereum Blockchain
- 6 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (1) , 45-48
Comprehensive Design Space Exploration for Graph Neural Network Aggregation on GPUs
Graph neural networks (GNNs) have become the state-of-the-art technology for extracting and predicting data representations on graphs. With increasing demand to accelerate GNN computations, the GPU has become the dominant platform for GNN training and inference. GNNs consist of a compute-bound combination phase and a memory-bound aggregation phase. The memory access patterns of the aggregation phase remain a major performance bottleneck on GPUs, despite recent microarchitectural enhancements. Although GNN characterizations have been conducted to investigate this bottleneck, they did not reveal the impact of architectural modifications. However, a comprehensive understanding of improvements from such modifications is imperative to devise GPU optimizations for the aggregation phase. In this letter, we explore the GPU design space for aggregation by assessing the performance improvement potential of a series of architectural modifications. We find that the low locality of aggregation deteriorates performance with increased thread-level parallelism, and a significant enhancement follows memory access optimizations, which remain effective even with software optimization. Our analysis provides insights for hardware optimizations to significantly improve GNN aggregation on GPUs.
- 6 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (1) , 45-48
Comprehensive Design Space Exploration for Graph Neural Network Aggregation on GPUs
Graph neural networks (GNNs) have become the state-of-the-art technology for extracting and predicting data representations on graphs. With increasing demand to accelerate GNN computations, the GPU has become the dominant platform for GNN training and inference. GNNs consist of a compute-bound combination phase and a memory-bound aggregation phase. The memory access patterns of the aggregation phase remain a major performance bottleneck on GPUs, despite recent microarchitectural enhancements. Although GNN characterizations have been conducted to investigate this bottleneck, they did not reveal the impact of architectural modifications. However, a comprehensive understanding of improvements from such modifications is imperative to devise GPU optimizations for the aggregation phase. In this letter, we explore the GPU design space for aggregation by assessing the performance improvement potential of a series of architectural modifications. We find that the low locality of aggregation deteriorates performance with increased thread-level parallelism, and a significant enhancement follows memory access optimizations, which remain effective even with software optimization. Our analysis provides insights for hardware optimizations to significantly improve GNN aggregation on GPUs.
- Open Access
BEC: Bit-Level Static Analysis for Reliability against Soft Errors
Soft errors are a type of transient digital signal corruption that occurs in digital hardware components such as the internal flip-flops of CPU pipelines, the register file, memory cells, and even internal communication buses. Soft errors are caused by environmental radioactivity, magnetic interference, lasers, and temperature fluctuations, either unintentionally, or as part of a deliberate attempt to compromise a system and expose confidential data. We propose a bit-level error coalescing (BEC) static program analysis and its two use cases to understand and improve program reliability against soft errors. The BEC analysis tracks each bit corruption in the register file and classifies the effect of the corruption by its semantics at compile time. The usefulness of the proposed analysis is demonstrated in two scenarios, fault injection campaign pruning, and reliability-aware program transformation. Experimental results show that bit-level analysis pruned up to 30.04 % of exhaustive fault injection campaigns (13.71 % on average), without loss of accuracy. Program vulnerability was reduced by up to 13.11 % (4.94 % on average) through bit-level vulnerability-aware instruction scheduling. The analysis has been implemented within LLVM and evaluated on the RISC-V architecture. To the best of our knowledge, the proposed BEC analysis is the first bit-level compiler analysis for program reliability against soft errors. The proposed method is generic and not limited to a specific computer architecture.
Open AccessBEC: Bit-Level Static Analysis for Reliability against Soft Errors
Soft errors are a type of transient digital signal corruption that occurs in digital hardware components such as the internal flip-flops of CPU pipelines, the register file, memory cells, and even internal communication buses. Soft errors are caused by environmental radioactivity, magnetic interference, lasers, and temperature fluctuations, either unintentionally, or as part of a deliberate attempt to compromise a system and expose confidential data. We propose a bit-level error coalescing (BEC) static program analysis and its two use cases to understand and improve program reliability against soft errors. The BEC analysis tracks each bit corruption in the register file and classifies the effect of the corruption by its semantics at compile time. The usefulness of the proposed analysis is demonstrated in two scenarios, fault injection campaign pruning, and reliability-aware program transformation. Experimental results show that bit-level analysis pruned up to 30.04 % of exhaustive fault injection campaigns (13.71 % on average), without loss of accuracy. Program vulnerability was reduced by up to 13.11 % (4.94 % on average) through bit-level vulnerability-aware instruction scheduling. The analysis has been implemented within LLVM and evaluated on the RISC-V architecture. To the best of our knowledge, the proposed BEC analysis is the first bit-level compiler analysis for program reliability against soft errors. The proposed method is generic and not limited to a specific computer architecture.
- Open Access
Julia Cloud Matrix Machine: Dynamic Matrix Language Acceleration on Multicore Clusters in the Cloud
In emerging scientific computing environments, matrix computations of increasing size and complexity are increasingly becoming prevalent. However, contemporary matrix language implementations are insufficient in their support for efficient utilization of cloud computing resources, particularly on the user side. We thus developed an extension of the Julia high-performance computation language such that matrix computations are automatically parallelized in the cloud, where users are separated from directly interacting with complex explicitly-parallel computations. We implement lazy evaluation semantics combined with directed graphs to optimize matrix operations on the fly while dynamic simulation finds the optimal tile size and schedule for a given cluster of cloud nodes. A time model prediction of the cluster's performance capacity is constructed to enable simulations. Automatic configuration of communication and worker processes on the cloud networks allow for the framework to automatically scale up for clusters of heterogeneous nodes. Our framework's experimental evaluation comprises eleven benchmarks on an fourteen node (564 CPUs) cluster in the AWS public cloud, revealing speedups of up to a factor of 5.1, with an average 74.39% of the upper bound for speedups.
Open AccessJulia Cloud Matrix Machine: Dynamic Matrix Language Acceleration on Multicore Clusters in the Cloud
In emerging scientific computing environments, matrix computations of increasing size and complexity are increasingly becoming prevalent. However, contemporary matrix language implementations are insufficient in their support for efficient utilization of cloud computing resources, particularly on the user side. We thus developed an extension of the Julia high-performance computation language such that matrix computations are automatically parallelized in the cloud, where users are separated from directly interacting with complex explicitly-parallel computations. We implement lazy evaluation semantics combined with directed graphs to optimize matrix operations on the fly while dynamic simulation finds the optimal tile size and schedule for a given cluster of cloud nodes. A time model prediction of the cluster's performance capacity is constructed to enable simulations. Automatic configuration of communication and worker processes on the cloud networks allow for the framework to automatically scale up for clusters of heterogeneous nodes. Our framework's experimental evaluation comprises eleven benchmarks on an fourteen node (564 CPUs) cluster in the AWS public cloud, revealing speedups of up to a factor of 5.1, with an average 74.39% of the upper bound for speedups.
- Open Access
- 17 October 2023
- Published by Springer Nature in Cluster Computing
- Vol. 27 (2) , 2309
Correction: Scalable NUMA-aware persistent B+-tree for non-volatile memory devices
- 17 October 2023
- Published by Springer Nature in Cluster Computing
- Vol. 27 (2) , 2309
Open AccessCorrection: Scalable NUMA-aware persistent B+-tree for non-volatile memory devices
- Open Access
- 14 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 47159-47178
EVMTracer: Dynamic Analysis of the Parallelization and Redundancy Potential in the Ethereum Virtual Machine
Ethereum is one of the first blockchains executing smart contracts (financial applications directly executed on the ledger using a virtual machine). The performance of the Ethereum virtual machine has become a bottleneck caused by high volumes of transactions due to blockchain applications, including decentralized finance and non-fungible tokens. Hence, there is a need to detail the execution characteristics of the Ethereum virtual machine before its performance can be improved. This work introduces an off-line Ethereum virtual machine tracer called EVMTracer that produces runtime data dependence graphs from executed transactions as an alternative program representation. From the runtime dependence graphs, we can deduce valuable metrics about program execution characteristics, including the degree of parallelization and computational redundancies in smart contracts. Our experiments encompass all blocks up to 12M on the Ethereum mainnet. We found a geometric mean of 1.90× theoretical maximum speedup when executing the smart contracts in parallel and identified 34.97% of SLOAD instructions as redundant.
- 14 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 47159-47178
Open AccessEVMTracer: Dynamic Analysis of the Parallelization and Redundancy Potential in the Ethereum Virtual Machine
Ethereum is one of the first blockchains executing smart contracts (financial applications directly executed on the ledger using a virtual machine). The performance of the Ethereum virtual machine has become a bottleneck caused by high volumes of transactions due to blockchain applications, including decentralized finance and non-fungible tokens. Hence, there is a need to detail the execution characteristics of the Ethereum virtual machine before its performance can be improved. This work introduces an off-line Ethereum virtual machine tracer called EVMTracer that produces runtime data dependence graphs from executed transactions as an alternative program representation. From the runtime dependence graphs, we can deduce valuable metrics about program execution characteristics, including the degree of parallelization and computational redundancies in smart contracts. Our experiments encompass all blocks up to 12M on the Ethereum mainnet. We found a geometric mean of 1.90× theoretical maximum speedup when executing the smart contracts in parallel and identified 34.97% of SLOAD instructions as redundant.