Heinrich Riebler
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Heinrich Riebler has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Christian Plessl
Paderborn University, Paderborn, Germany
14 shared publications Gavin Vaz
Paderborn Center for Parallel Computing (PC2), Paderborn University, Paderborn, Germany
6 shared publicationsTobias Kenter
Paderborn Center for Parallel Computing (PC2), Paderborn University, Paderborn, Germany
5 shared publications- Bjarne Wintermann
Paderborn University, Paderborn Center for Parallel Computing, Paderborn, Germany
4 shared publications - Linus Jungemann
Paderborn University, Paderborn Center for Parallel Computing, Paderborn, Germany
4 shared publications - Michael Lass
Department of Computer Science Paderborn University
3 shared publications
Latest Publications
- Open Access
- 25 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 35-48
AuroraFlow, an Easy-to-Use, Low-Latency FPGA Communication Solution Demonstrated on Multi-FPGA Neural Network Inference
- 25 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 35-48
Open AccessAuroraFlow, an Easy-to-Use, Low-Latency FPGA Communication Solution Demonstrated on Multi-FPGA Neural Network Inference
- Open Access
- 25 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 103-116
FINN-HPC: Closing the Gap for Energy-Efficient Neural Network Inference on FPGAs in HPC
- 25 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 103-116
Open AccessFINN-HPC: Closing the Gap for Energy-Efficient Neural Network Inference on FPGAs in HPC
- Open Access
- 27 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 52
Neural Network Inference in High-Performance Computing: Closing the Gap for FINN based Reconfigurable Accelerators
In recent years, Neural Networks (NNs) have become one of the most prevailing topics in computers science, both in research and in industry. NNs are used for data analysis, natural language processing, autonomous driving and more. As such, NNs also see more application and use in High-Performance Computing (HPC). At the same time, energy efficiency has become an increasingly critical topic. NNs use large amounts of energy for operation, which in return results in large amounts of CO2 emissions. This work presents a comprehensive evaluation of current NN inference soft- and hardware configurations within High-Performance Computing (HPC) environments, with a focus on both performance metrics and energy consumption. NN quantization and accelerators such as FPGAs allow for an increased inference efficiency, both in terms of throughput and energy. Therefore, this work focuses on FINN, an efficient NN inference framework for FPGAs, highlighting its current lack of support for HPC systems. We provide an in-depth analysis of FINN in order to implement extensions to optimize the end-to-end execution for the usage in the HPC environment. We thoroughly evaluate the performance and energy efficiency gains using newly implemented optimizations and compare it against existing NN accelerators for HPC. With our extensions of FINN, we were able to achieve a 1847× higher throughput, while also decreasing the latency on average by 0.9978× and EDP by 0.9979× on an Alveo U55C FPGA. Data flow based NN inference accelerators on an FPGA should be used if the performance and energy footprint of the inference process is crucial, and the batch sizes are small to medium. For extremely large batch sizes and a very limited time for network-to-accelerator (less than a few days), using GPUs is still the way to go. Our results show that with the newly developed driver, we outperform a high-end Nvidia A100 GPU by up to 7.81x in throughput, while having a 0.87x lower latency and 0.88x lower energy delay product.
- 27 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 52
Open AccessNeural Network Inference in High-Performance Computing: Closing the Gap for FINN based Reconfigurable Accelerators
In recent years, Neural Networks (NNs) have become one of the most prevailing topics in computers science, both in research and in industry. NNs are used for data analysis, natural language processing, autonomous driving and more. As such, NNs also see more application and use in High-Performance Computing (HPC). At the same time, energy efficiency has become an increasingly critical topic. NNs use large amounts of energy for operation, which in return results in large amounts of CO2 emissions. This work presents a comprehensive evaluation of current NN inference soft- and hardware configurations within High-Performance Computing (HPC) environments, with a focus on both performance metrics and energy consumption. NN quantization and accelerators such as FPGAs allow for an increased inference efficiency, both in terms of throughput and energy. Therefore, this work focuses on FINN, an efficient NN inference framework for FPGAs, highlighting its current lack of support for HPC systems. We provide an in-depth analysis of FINN in order to implement extensions to optimize the end-to-end execution for the usage in the HPC environment. We thoroughly evaluate the performance and energy efficiency gains using newly implemented optimizations and compare it against existing NN accelerators for HPC. With our extensions of FINN, we were able to achieve a 1847× higher throughput, while also decreasing the latency on average by 0.9978× and EDP by 0.9979× on an Alveo U55C FPGA. Data flow based NN inference accelerators on an FPGA should be used if the performance and energy footprint of the inference process is crucial, and the batch sizes are small to medium. For extremely large batch sizes and a very limited time for network-to-accelerator (less than a few days), using GPUs is still the way to go. Our results show that with the newly developed driver, we outperform a high-end Nvidia A100 GPU by up to 7.81x in throughput, while having a 0.87x lower latency and 0.88x lower energy delay product.
- Open Access
- 30 September 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 17 (3) , 1-28
A Computation of the Ninth Dedekind Number Using FPGA Supercomputing
This manuscript makes the claim of having computed the \(9\) th Dedekind number, D(9). This was done by accelerating the core operation of the process with an efficient FPGA design that outperforms an optimized 64-core CPU reference by 95 \(\times\) . The FPGA execution was parallelized on the Noctua 2 supercomputer at Paderborn University. The resulting value for D(9) is 286386577668298411128469151667598498812366. This value can be verified in two steps. We have made the data file containing the 490 M results available, each of which can be verified separately on CPU, and the whole file sums to our proposed value. The paper explains the mathematical approach in the first part, before putting the focus on a deep dive into the FPGA accelerator implementation followed by a performance analysis. The FPGA implementation was done in Register-Transfer Level using a dual-clock architecture and shows how we achieved an impressive FMax of 450 MHz on the targeted Stratix 10 GX 2,800 FPGAs. The total compute time used was 47,000 FPGA hours.- 30 September 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 17 (3) , 1-28
Open AccessA Computation of the Ninth Dedekind Number Using FPGA Supercomputing
This manuscript makes the claim of having computed the \(9\) th Dedekind number, D(9). This was done by accelerating the core operation of the process with an efficient FPGA design that outperforms an optimized 64-core CPU reference by 95 \(\times\) . The FPGA execution was parallelized on the Noctua 2 supercomputer at Paderborn University. The resulting value for D(9) is 286386577668298411128469151667598498812366. This value can be verified in two steps. We have made the data file containing the 490 M results available, each of which can be verified separately on CPU, and the whole file sums to our proposed value. The paper explains the mathematical approach in the first part, before putting the focus on a deep dive into the FPGA accelerator implementation followed by a performance analysis. The FPGA implementation was done in Register-Transfer Level using a dual-clock architecture and shows how we achieved an impressive FMax of 450 MHz on the targeted Stratix 10 GX 2,800 FPGAs. The total compute time used was 47,000 FPGA hours. - Open Access
- 18 April 2019
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 16 (2) , 1-26
Transparent Acceleration for Heterogeneous Platforms With Compilation to OpenCL
Multi-accelerator platforms combine CPUs and different accelerator architectures within a single compute node. Such systems are capable of processing parallel workloads very efficiently while being more energy efficient than regular systems consisting of CPUs only. However, the architectures of such systems are diverse, forcing developers to port applications to each accelerator using different programming languages, models, tools, and compilers. Developers not only require domain-specific knowledge but also need to understand the low-level accelerator details, leading to an increase in the design effort and costs. To tackle this challenge, we propose a compilation approach and a practical realization called HT r OP that is completely transparent to the user. HT r OP is able to automatically analyze a sequential CPU application, detect computational hotspots, and generate parallel OpenCL host and kernel code. The potential of HT r OP is demonstrated by offloading hotspots to different OpenCL-enabled resources (currently the CPU, the general-purpose GPU, and the manycore Intel Xeon Phi) for a broad set of benchmark applications. We present an in-depth evaluation of our approach in terms of performance gains and energy savings, taking into account all static and dynamic overheads. We are able to achieve speedups and energy savings of up to two orders of magnitude, if an application has sufficient computational intensity, when compared to a natively compiled application.
- 18 April 2019
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 16 (2) , 1-26
Open AccessTransparent Acceleration for Heterogeneous Platforms With Compilation to OpenCL
Multi-accelerator platforms combine CPUs and different accelerator architectures within a single compute node. Such systems are capable of processing parallel workloads very efficiently while being more energy efficient than regular systems consisting of CPUs only. However, the architectures of such systems are diverse, forcing developers to port applications to each accelerator using different programming languages, models, tools, and compilers. Developers not only require domain-specific knowledge but also need to understand the low-level accelerator details, leading to an increase in the design effort and costs. To tackle this challenge, we propose a compilation approach and a practical realization called HT r OP that is completely transparent to the user. HT r OP is able to automatically analyze a sequential CPU application, detect computational hotspots, and generate parallel OpenCL host and kernel code. The potential of HT r OP is demonstrated by offloading hotspots to different OpenCL-enabled resources (currently the CPU, the general-purpose GPU, and the manycore Intel Xeon Phi) for a broad set of benchmark applications. We present an in-depth evaluation of our approach in terms of performance gains and energy savings, taking into account all static and dynamic overheads. We are able to achieve speedups and energy savings of up to two orders of magnitude, if an application has sufficient computational intensity, when compared to a natively compiled application.
- Open Access
- 10 February 2018
- Published by Association for Computing Machinery (ACM)
- p. 417-418
Automated code acceleration targeting heterogeneous openCL devices
Accelerators can offer exceptional performance advantages. However, programmers need to spend considerable efforts on acceleration, without knowing how sustainable the employed programming models, languages and tools are. To tackle this challenge, we propose and demonstrate a new runtime system called HTrOP that is able to automatically generate and execute OpenCL code from sequential CPU code. HTrOP transforms suitable data-parallel loops into independent OpenCL-typical work-items and handles concrete calls to these devices through a mix of library components and application-specific OpenCL host code. Computational hotspots are identified and can be offloaded to different resources (CPU, GPGPU and Xeon Phi). We demonstrate the potential of HTrOP on a broad set of applications and are able to improve the performance by 4.3X on average.
- 10 February 2018
- Published by Association for Computing Machinery (ACM)
- p. 417-418
Open AccessAutomated code acceleration targeting heterogeneous openCL devices
Accelerators can offer exceptional performance advantages. However, programmers need to spend considerable efforts on acceleration, without knowing how sustainable the employed programming models, languages and tools are. To tackle this challenge, we propose and demonstrate a new runtime system called HTrOP that is able to automatically generate and execute OpenCL code from sequential CPU code. HTrOP transforms suitable data-parallel loops into independent OpenCL-typical work-items and handles concrete calls to these devices through a mix of library components and application-specific OpenCL host code. Computational hotspots are identified and can be offloaded to different resources (CPU, GPGPU and Xeon Phi). We demonstrate the potential of HTrOP on a broad set of applications and are able to improve the performance by 4.3X on average.
- Open Access
- 27 June 2017
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 10 (3) , 1-23
Efficient Branch and Bound on FPGAs Using Work Stealing and Instance-Specific Designs
Branch and bound (B8B) algorithms structure the search space as a tree and eliminate infeasible solutions early by pruning subtrees that cannot lead to a valid or optimal solution. Custom hardware designs significantly accelerate the execution of these algorithms. In this article, we demonstrate a high-performance B8B implementation on FPGAs. First, we identify general elements of B8B algorithms and describe their implementation as a finite state machine. Then, we introduce workers that autonomously cooperate using work stealing to allow parallel execution and full utilization of the target FPGA. Finally, we explore advantages of instance-specific designs that target a specific problem instance to improve performance. We evaluate our concepts by applying them to a branch and bound problem, the reconstruction of corrupted AES keys obtained from cold-boot attacks. The evaluation shows that our work stealing approach is scalable with the available resources and provides speedups proportional to the number of workers. Instance-specific designs allow us to achieve an overall speedup of 47 × compared to the fastest implementation of AES key reconstruction so far. Finally, we demonstrate how instance-specific designs can be generated just-in-time such that the provided speedups outweigh the additional time required for design synthesis.
- 27 June 2017
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
- Vol. 10 (3) , 1-23
Open AccessEfficient Branch and Bound on FPGAs Using Work Stealing and Instance-Specific Designs
Branch and bound (B8B) algorithms structure the search space as a tree and eliminate infeasible solutions early by pruning subtrees that cannot lead to a valid or optimal solution. Custom hardware designs significantly accelerate the execution of these algorithms. In this article, we demonstrate a high-performance B8B implementation on FPGAs. First, we identify general elements of B8B algorithms and describe their implementation as a finite state machine. Then, we introduce workers that autonomously cooperate using work stealing to allow parallel execution and full utilization of the target FPGA. Finally, we explore advantages of instance-specific designs that target a specific problem instance to improve performance. We evaluate our concepts by applying them to a branch and bound problem, the reconstruction of corrupted AES keys obtained from cold-boot attacks. The evaluation shows that our work stealing approach is scalable with the available resources and provides speedups proportional to the number of workers. Instance-specific designs allow us to achieve an overall speedup of 47 × compared to the fastest implementation of AES key reconstruction so far. Finally, we demonstrate how instance-specific designs can be generated just-in-time such that the provided speedups outweigh the additional time required for design synthesis.
- 14 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Using just-in-time code generation for transparent resource management in heterogeneous systems
Hardware accelerators are becoming popular in academia and industry. To move one step further from the state-of-the-art multicore plus accelerator approaches, we present in this paper our innovative SAVEHSA architecture. It comprises of a heterogeneous hardware platform with three different high-end accelerators attached over PCIe (GPGPU, FPGA and Intel MIC). Such systems can process parallel workloads very efficiently whilst being more energy efficient than regular CPU systems. To leverage the heterogeneity, the workload has to be distributed among the computing units in a way that each unit is well-suited for the assigned task and executable code must be available. To tackle this problem we present two software components; the first can perform resource allocation at runtime while respecting system and application goals (in terms of throughput, energy, latency, etc.) and the second is able to analyze an application and generate executable code for an accelerator at runtime. We demonstrate the first proof-of-concept implementation of our framework on the heterogeneous platform, discuss different runtime policies and measure the introduced overheads.
- 14 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Using just-in-time code generation for transparent resource management in heterogeneous systems
Hardware accelerators are becoming popular in academia and industry. To move one step further from the state-of-the-art multicore plus accelerator approaches, we present in this paper our innovative SAVEHSA architecture. It comprises of a heterogeneous hardware platform with three different high-end accelerators attached over PCIe (GPGPU, FPGA and Intel MIC). Such systems can process parallel workloads very efficiently whilst being more energy efficient than regular CPU systems. To leverage the heterogeneity, the workload has to be distributed among the computing units in a way that each unit is well-suited for the assigned task and executable code must be available. To tackle this problem we present two software components; the first can perform resource allocation at runtime while respecting system and application goals (in terms of throughput, energy, latency, etc.) and the second is able to analyze an application and generate executable code for an accelerator at runtime. We demonstrate the first proof-of-concept implementation of our framework on the heterogeneous platform, discuss different runtime policies and measure the introduced overheads.
- 1 October 2016
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 55, 91-111
Potential and methods for embedding dynamic offloading decisions into application code
- 1 October 2016
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 55, 91-111
Potential and methods for embedding dynamic offloading decisions into application code