Clemens Grelck
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Clemens Grelck has a total of 83 co-authors. Here are the top co-authors by number of shared publications:
- Andy D. Pimentel
University of Amsterdam
3 shared publications - Lukas Miedema
University of Amsterdam
3 shared publications - Christoph Kessler
Linköping University
2 shared publications - Cosmin Eugen Oancea
DIKU, University of Copenhagen
2 shared publications - Elisabeth Larsson
Uppsala University
2 shared publications - Julius Roeder
University of Amsterdam
2 shared publications
Latest Publications
- Open Access
- 18 November 2026
- Published by Elsevier in Journal of Parallel and Distributed Computing
Comparing Parallel Functional Array Languages: Programming and Performance
- 18 November 2026
- Published by Elsevier in Journal of Parallel and Distributed Computing
Open AccessComparing Parallel Functional Array Languages: Programming and Performance
- Open Access
Energy-Efficient QoS-Aware Scheduling for S-NUCA Many-Cores
Optimizing performance and energy efficiency in many-core processors, especially within Non-Uniform Cache Access (NUCA) architectures, remains a critical challenge. The performance heterogeneity inherent in S-NUCA systems complicates task scheduling due to varying cache access latencies across cores. This paper introduces a novel QoS management policy to maintain application execution within predefined Quality of Service (QoS) targets, measured using the Application Heartbeats framework. QoS metrics like Heartbeats ensure predictable application performance in dynamic computing environments. The proposed policy dynamically controls QoS by orchestrating task migrations within the S-NUCA many-core system and adjusting the clock frequency of cores. After satisfying the QoS objectives, the policy optimizes energy efficiency, reducing overall system energy consumption without compromising performance constraints. Our work leverages the state-of-the-art multi-/many-core simulator {\em HotSniper}. We have extended it with two key components: an integrated heartbeat framework for precise, application-specific performance monitoring, and our QoS management policy that maintains application QoS requirements while minimizing the system's energy consumption. Experimental evaluations demonstrate that our approach effectively maintains desired QoS levels and achieves 18.7\% energy savings compared to state-of-the-art scheduling methods.
Open AccessEnergy-Efficient QoS-Aware Scheduling for S-NUCA Many-Cores
Optimizing performance and energy efficiency in many-core processors, especially within Non-Uniform Cache Access (NUCA) architectures, remains a critical challenge. The performance heterogeneity inherent in S-NUCA systems complicates task scheduling due to varying cache access latencies across cores. This paper introduces a novel QoS management policy to maintain application execution within predefined Quality of Service (QoS) targets, measured using the Application Heartbeats framework. QoS metrics like Heartbeats ensure predictable application performance in dynamic computing environments. The proposed policy dynamically controls QoS by orchestrating task migrations within the S-NUCA many-core system and adjusting the clock frequency of cores. After satisfying the QoS objectives, the policy optimizes energy efficiency, reducing overall system energy consumption without compromising performance constraints. Our work leverages the state-of-the-art multi-/many-core simulator {\em HotSniper}. We have extended it with two key components: an integrated heartbeat framework for precise, application-specific performance monitoring, and our QoS management policy that maintains application QoS requirements while minimizing the system's energy consumption. Experimental evaluations demonstrate that our approach effectively maintains desired QoS levels and achieves 18.7\% energy savings compared to state-of-the-art scheduling methods.
- Open Access
Comparing Parallel Functional Array Languages: Programming and Performance
Parallel functional array languages are an emerging class of programming languages that promise to combine low-effort parallel programming with good performance and performance portability. We systematically compare the designs and implementations of five different functional array languages: Accelerate, APL, DaCe, Futhark, and SaC. We demonstrate the expressiveness of functional array programming by means of four challenging benchmarks, namely N-body simulation, MultiGrid, Quickhull, and Flash Attention. These benchmarks represent a range of application domains and parallel computational models. We argue that the functional array code is much shorter and more comprehensible than the hand-optimized baseline implementations because it omits architecture-specific aspects. Instead, the language implementations generate both multicore and GPU executables from a single source code base. Hence, we further argue that functional array code could more easily be ported to, and optimized for, new parallel architectures than conventional implementations of numerical kernels. We demonstrate this potential by reporting the performance of the five parallel functional array languages on a total of 39 instances of the four benchmarks on both a 32-core AMD EPYC 7313 multicore system and on an NVIDIA A30 GPU. We explore in-depth why each language performs well or not so well on each benchmark and architecture. We argue that the results demonstrate that mature functional array languages have the potential to deliver performance competitive with the best available conventional techniques.
Open AccessComparing Parallel Functional Array Languages: Programming and Performance
Parallel functional array languages are an emerging class of programming languages that promise to combine low-effort parallel programming with good performance and performance portability. We systematically compare the designs and implementations of five different functional array languages: Accelerate, APL, DaCe, Futhark, and SaC. We demonstrate the expressiveness of functional array programming by means of four challenging benchmarks, namely N-body simulation, MultiGrid, Quickhull, and Flash Attention. These benchmarks represent a range of application domains and parallel computational models. We argue that the functional array code is much shorter and more comprehensible than the hand-optimized baseline implementations because it omits architecture-specific aspects. Instead, the language implementations generate both multicore and GPU executables from a single source code base. Hence, we further argue that functional array code could more easily be ported to, and optimized for, new parallel architectures than conventional implementations of numerical kernels. We demonstrate this potential by reporting the performance of the five parallel functional array languages on a total of 39 instances of the four benchmarks on both a 32-core AMD EPYC 7313 multicore system and on an NVIDIA A30 GPU. We explore in-depth why each language performs well or not so well on each benchmark and architecture. We argue that the results demonstrate that mature functional array languages have the potential to deliver performance competitive with the best available conventional techniques.
- 28 January 2025
- Published by Springer Nature
- p. 1-17
FAA+RTS: Designing Fault-Aware Adaptive Real-Time Systems —From Specification to Execution—
- 28 January 2025
- Published by Springer Nature
- p. 1-17
FAA+RTS: Designing Fault-Aware Adaptive Real-Time Systems —From Specification to Execution—
- 18 October 2023
- Published by Springer Nature
- p. 358-374
Code Quality Metrics for Functional Features in Modern Object-Oriented Languages
- 18 October 2023
- Published by Springer Nature
- p. 358-374
Code Quality Metrics for Functional Features in Modern Object-Oriented Languages
- 6 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 452-461
Change of Plans: Optimizing for Power, Reliability and Timeliness for Cost-Conscious Real-Time Systems
The large-scale adoption of embedded devices naturally sees them deployed under widely varying constraints matching an equally wide number of environments and applications. Such devices may be under hard timing constraints as a result of their domain (real-time), but other constraints, such as reliability and power, are becoming increasingly relevant. In particular, battery-powered devices are limited in runtime due to their long-running average power (amortized power) consumption. We propose a real-time scheduling algorithm optimizing for reliability under an amortized energy budget. Weakly-hard real-time tasks can tolerate patterns of failures and successes, and we leverage this property as the driver of reliability in a constrained environment. Reliability against transient faults is increased by dynamically switching the protected subset of the task set based on the observed history of task success/fails. By performing such a change of plans only for tasks in need, we can cost-effectively boost reliability while lowering energy consumption. Our method optimizes reliability for a given deadline and energy budget. We explore the behavior of our method under a wide range of energy and time constrained environments, illustrating the trade-off between amortized power, timeliness and reliability, enabling application designers to make cost-conscious decisions.
- 6 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 452-461
Change of Plans: Optimizing for Power, Reliability and Timeliness for Cost-Conscious Real-Time Systems
The large-scale adoption of embedded devices naturally sees them deployed under widely varying constraints matching an equally wide number of environments and applications. Such devices may be under hard timing constraints as a result of their domain (real-time), but other constraints, such as reliability and power, are becoming increasingly relevant. In particular, battery-powered devices are limited in runtime due to their long-running average power (amortized power) consumption. We propose a real-time scheduling algorithm optimizing for reliability under an amortized energy budget. Weakly-hard real-time tasks can tolerate patterns of failures and successes, and we leverage this property as the driver of reliability in a constrained environment. Reliability against transient faults is increased by dynamically switching the protected subset of the task set based on the observed history of task success/fails. By performing such a change of plans only for tasks in need, we can cost-effectively boost reliability while lowering energy consumption. Our method optimizes reliability for a given deadline and energy budget. We explore the behavior of our method under a wide range of energy and time constrained environments, illustrating the trade-off between amortized power, timeliness and reliability, enabling application designers to make cost-conscious decisions.
- Open Access
- 29 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Fault-tolerance at your Finger Tips with the TeamPlay Coordination Language
- 29 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessFault-tolerance at your Finger Tips with the TeamPlay Coordination Language
- 1 June 2023
- Published by Springer Nature
- p. 121-134
GCN-based Reinforcement Learning Approach for Scheduling DAG Applications
- 1 June 2023
- Published by Springer Nature
- p. 121-134
GCN-based Reinforcement Learning Approach for Scheduling DAG Applications
- 1 October 2022
- Published by Springer Nature
- p. 129-145
Strategy Switching: Smart Fault-Tolerance for Weakly-Hard Resource-Constrained Real-Time Applications
- 1 October 2022
- Published by Springer Nature
- p. 129-145
Strategy Switching: Smart Fault-Tolerance for Weakly-Hard Resource-Constrained Real-Time Applications
- 10 June 2020
- Published by Springer Nature
- Vol. 12134, 57-74
Towards Energy-, Time- and Security-Aware Multi-core Coordination
Coordination is a well established computing paradigm with a plethora of languages, abstractions and approaches. Yet, we are not aware of any adoption of the principle of coordination in the broad domain of cyber-physical systems, where non-functional properties, such as execution/response time, energy consumption and security are as crucial as functional correctness. We propose a coordination approach, including a functional coordination language and its associated tool flow, that considers time, energy and security as first-class citizens in application design and development. We primarily target cyber-physical systems running on off-the-shelf heterogeneous multi-core platforms. We illustrate our approach by means of a real-world use case, an unmanned aerial vehicle for autonomous reconnaissance mission, which we develop in close collaboration with industry.
- 10 June 2020
- Published by Springer Nature
- Vol. 12134, 57-74
Towards Energy-, Time- and Security-Aware Multi-core Coordination
Coordination is a well established computing paradigm with a plethora of languages, abstractions and approaches. Yet, we are not aware of any adoption of the principle of coordination in the broad domain of cyber-physical systems, where non-functional properties, such as execution/response time, energy consumption and security are as crucial as functional correctness. We propose a coordination approach, including a functional coordination language and its associated tool flow, that considers time, energy and security as first-class citizens in application design and development. We primarily target cyber-physical systems running on off-the-shelf heterogeneous multi-core platforms. We illustrate our approach by means of a real-world use case, an unmanned aerial vehicle for autonomous reconnaissance mission, which we develop in close collaboration with industry.