Georgios Goumas
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Georgios Goumas has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Nectarios Koziris
National Technical University of Athens
31 shared publications - Konstantinos Nikas
Computing Systems Laboratory, National Technical University of Athens, 15772 Zografou, Greece
10 shared publications - Constantinos Bitsakos
Computing Systems Laboratory, National Technical University of Athens, 15772 Zografou, Greece
7 shared publications - Theodore Andronikos
Department of Informatics, Ionian University, 7 Tsirigoti Square, 49100 Corfu, Greece
7 shared publications - Nikela Papadopoulou
University of Glasgow, Glasgow, United Kingdom
5 shared publications - Chloe Alverti
University of Illinois Urbana-Champaign, Urbana-Champaign, Illinois, USA
4 shared publications
Latest Publications
- Open Access
- 26 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23 (4) , 1-25
Demystifying SpMV Performance: Bottleneck Characterization and Open Challenges
Sparse Matrix-Vector Multiplication (SpMV) is a critical computational kernel in scientific computing, graph processing, and machine learning. Despite its importance, SpMV achieves only a small fraction of peak performance, varying substantially across matrices and architectures. This paper comprehensively analyzes SpMV performance on modern CPUs and GPUs, identifying key bottlenecks: memory bandwidth limitations, low instruction-level parallelism (ILP), load imbalance, and memory latency overheads. We characterize matrices via four structural features, each linked to a bottleneck, to enable structured performance analysis. We also develop and publicly release a matrix generator producing artificial matrices across a broad range of structural properties, enabling extensive benchmarking from high-end HPC accelerators to consumer-grade GPUs. While GPUs achieve superior SpMV performance, modern CPUs with larger caches and increased core counts remain competitive, particularly for small-to-medium sized matrices. Evaluating numerous storage formats, we demonstrate the efficiency of vendor-customized implementations on GPUs and the potential of research formats to address specific CPU performance challenges. Expanding to iterative solvers (BiCG) and FP32 arithmetic reveals non-obvious architectural asymmetries: GPU throughput is resilient to solver-level cache pressure while CPUs degrade significantly, and FP32 scaling yields architecture-specific gains, from bandwidth improvements to unlocked compute capacity on consumer GPUs. Finally, we identify persistent open challenges: memory bandwidth constraints, irregular matrix structures, and short unbalanced rows, underscoring the need for further optimization and architectural improvements.
- 26 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 23 (4) , 1-25
Open AccessDemystifying SpMV Performance: Bottleneck Characterization and Open Challenges
Sparse Matrix-Vector Multiplication (SpMV) is a critical computational kernel in scientific computing, graph processing, and machine learning. Despite its importance, SpMV achieves only a small fraction of peak performance, varying substantially across matrices and architectures. This paper comprehensively analyzes SpMV performance on modern CPUs and GPUs, identifying key bottlenecks: memory bandwidth limitations, low instruction-level parallelism (ILP), load imbalance, and memory latency overheads. We characterize matrices via four structural features, each linked to a bottleneck, to enable structured performance analysis. We also develop and publicly release a matrix generator producing artificial matrices across a broad range of structural properties, enabling extensive benchmarking from high-end HPC accelerators to consumer-grade GPUs. While GPUs achieve superior SpMV performance, modern CPUs with larger caches and increased core counts remain competitive, particularly for small-to-medium sized matrices. Evaluating numerous storage formats, we demonstrate the efficiency of vendor-customized implementations on GPUs and the potential of research formats to address specific CPU performance challenges. Expanding to iterative solvers (BiCG) and FP32 arithmetic reveals non-obvious architectural asymmetries: GPU throughput is resilient to solver-level cache pressure while CPUs degrade significantly, and FP32 scaling yields architecture-specific gains, from bandwidth improvements to unlocked compute capacity on consumer GPUs. Finally, we identify persistent open challenges: memory bandwidth constraints, irregular matrix structures, and short unbalanced rows, underscoring the need for further optimization and architectural improvements.
- Open Access
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 279-289
Low-Latency ML Offloading Across Edge and IoT Devices
Deploying machine learning (ML) inference across the Cloud\allowbreak–\allowbreak Edge\allowbreak–\allowbreak IoT continuum faces challenges from device heterogeneity, limited resources, and dynamic networks. This work presents MLIoT, a cloud-native framework enabling seamless ML workload offloading with secure, dynamic device discovery. Building on Akri and vAccel, MLIoT balances local inference on constrained microcontrollers with offloading to edge accelerators, optimizing latency, resource use, and accuracy. In this work, we highlight the trade-offs between performing inference locally and transparently offloading compute-intensive components to edge accelerators. Large models often cannot fit in constrained memory, and quantized local models can suffer accuracy loss. Similarly, compute-limited devices may experience high latency or energy consumption when running inference locally. We describe performance engineering techniques including custom discovery, transport optimizations, and scheduling. Experiments on ESP32, Raspberry Pi, NVIDIA Jetson, and x86 cloud nodes demonstrate reduced latency and improved inference efficiency, highlighting practical trade-offs and challenges in managing dynamic device availability. Our findings showcase the potential of cloud-native performance engineering for accurate, low-latency ML inference at the edge.
- 3 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 279-289
Open AccessLow-Latency ML Offloading Across Edge and IoT Devices
Deploying machine learning (ML) inference across the Cloud\allowbreak–\allowbreak Edge\allowbreak–\allowbreak IoT continuum faces challenges from device heterogeneity, limited resources, and dynamic networks. This work presents MLIoT, a cloud-native framework enabling seamless ML workload offloading with secure, dynamic device discovery. Building on Akri and vAccel, MLIoT balances local inference on constrained microcontrollers with offloading to edge accelerators, optimizing latency, resource use, and accuracy. In this work, we highlight the trade-offs between performing inference locally and transparently offloading compute-intensive components to edge accelerators. Large models often cannot fit in constrained memory, and quantized local models can suffer accuracy loss. Similarly, compute-limited devices may experience high latency or energy consumption when running inference locally. We describe performance engineering techniques including custom discovery, transport optimizations, and scheduling. Experiments on ESP32, Raspberry Pi, NVIDIA Jetson, and x86 cloud nodes demonstrate reduced latency and improved inference efficiency, highlighting practical trade-offs and challenges in managing dynamic device availability. Our findings showcase the potential of cloud-native performance engineering for accurate, low-latency ML inference at the edge.
- Open Access
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1467-1482
Squeezy: Rapid VM Memory Reclamation for Serverless Functions
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1467-1482
Open AccessSqueezy: Rapid VM Memory Reclamation for Serverless Functions
- 2 January 2026
- Published by Springer Nature
- p. 284-305
Performance Models to Support HPC Co-scheduling
- 2 January 2026
- Published by Springer Nature
- p. 306-326
ELiSE: A Tool to Support Algorithmic Design for HPC Co-scheduling
- 2 January 2026
- Published by Springer Nature
- p. 306-326
ELiSE: A Tool to Support Algorithmic Design for HPC Co-scheduling
- Open Access
Probabilistic Links Between Quantum Classification of Patterns of Boolean Functions and Hamming Distance
This article investigates the probabilistic relationship between quantum classification of Boolean functions and their Hamming distance. By integrating concepts from quantum computing, information theory, and combinatorics, we explore how Hamming distance serves as a metric for analyzing deviations in function classification. Our extensive experimental results confirm that the Hamming distance is a pivotal metric for validating nearest neighbors in the process of classifying random functions. One of the significant conclusions we arrived is that the successful classification probability decreases monotonically with the Hamming distance. However, key exceptions were found in specific classes, revealing intra-class heterogeneity. We have established that these deviations are not random but are systemic and predictable. Furthermore, we were able to quantify these irregularities, turning potential errors into manageable phenomena. The most important novelty of this work is the demarcation, for the first time to the best of our knowledge, of precise Hamming distance intervals for the classification probability. These intervals bound the possible values the probability can assume, and provide a new foundational tool for probabilistic assessment in quantum classification. Practitioners can now endorse classification results with high certainty or dismiss them with confidence. This framework can significantly enhance any quantum classification algorithm’s reliability and decision-making capability.
Open AccessProbabilistic Links Between Quantum Classification of Patterns of Boolean Functions and Hamming Distance
This article investigates the probabilistic relationship between quantum classification of Boolean functions and their Hamming distance. By integrating concepts from quantum computing, information theory, and combinatorics, we explore how Hamming distance serves as a metric for analyzing deviations in function classification. Our extensive experimental results confirm that the Hamming distance is a pivotal metric for validating nearest neighbors in the process of classifying random functions. One of the significant conclusions we arrived is that the successful classification probability decreases monotonically with the Hamming distance. However, key exceptions were found in specific classes, revealing intra-class heterogeneity. We have established that these deviations are not random but are systemic and predictable. Furthermore, we were able to quantify these irregularities, turning potential errors into manageable phenomena. The most important novelty of this work is the demarcation, for the first time to the best of our knowledge, of precise Hamming distance intervals for the classification probability. These intervals bound the possible values the probability can assume, and provide a new foundational tool for probabilistic assessment in quantum classification. Practitioners can now endorse classification results with high certainty or dismiss them with confidence. This framework can significantly enhance any quantum classification algorithm’s reliability and decision-making capability.
- Open Access
Sparse Computations in Deep Learning Inference
The computational demands of modern Deep Neural Networks (DNNs) are immense and constantly growing. While training costs usually capture public attention, inference demands are also contributing in significant computational, energy and environmental footprints. Sparsity stands out as a critical mechanism for drastically reducing these resource demands. However, its potential remains largely untapped and is not yet fully incorporated in production AI systems. To bridge this gap, this work provides the necessary knowledge and insights for performance engineers keen to get involved in deep learning inference optimization. In particular, in this work we: a) discuss the various forms of sparsity that can be utilized in DNN inference, b) explain how the original dense computations translate to sparse kernels, c) provide an extensive bibliographic review of the state-of-the-art in the implementation of these kernels for CPUs and GPUs, d) discuss the availability of sparse datasets in support of sparsity-related research and development, e) explore the current software tools and frameworks that provide robust sparsity support, and f) present evaluation results of different implementations of the key SpMM and SDDMM kernels on CPU and GPU platforms. Ultimately, this paper aims to serve as a resource for performance engineers seeking to develop and deploy highly efficient sparse deep learning models in productions.
Open AccessSparse Computations in Deep Learning Inference
The computational demands of modern Deep Neural Networks (DNNs) are immense and constantly growing. While training costs usually capture public attention, inference demands are also contributing in significant computational, energy and environmental footprints. Sparsity stands out as a critical mechanism for drastically reducing these resource demands. However, its potential remains largely untapped and is not yet fully incorporated in production AI systems. To bridge this gap, this work provides the necessary knowledge and insights for performance engineers keen to get involved in deep learning inference optimization. In particular, in this work we: a) discuss the various forms of sparsity that can be utilized in DNN inference, b) explain how the original dense computations translate to sparse kernels, c) provide an extensive bibliographic review of the state-of-the-art in the implementation of these kernels for CPUs and GPUs, d) discuss the availability of sparse datasets in support of sparsity-related research and development, e) explore the current software tools and frameworks that provide robust sparsity support, and f) present evaluation results of different implementations of the key SpMM and SDDMM kernels on CPU and GPU platforms. Ultimately, this paper aims to serve as a resource for performance engineers seeking to develop and deploy highly efficient sparse deep learning models in productions.
- Open Access
Probabilistic Links Between Quantum Classification of Patterns of Boolean Functions and Hamming Distance
This article investigates the probabilistic relationship between quantum classification of Boolean functions and their Hamming distance. By integrating concepts from quantum computing, information theory, and combinatorics, we explore how Hamming distance serves as a metric for analyzing deviations in function classification. Our extensive experimental results confirm that the Hamming distance is a pivotal metric for validating nearest neighbors in the process of classifying random functions. One of the significant conclusions we arrived is that the successful classification probability decreases monotonically with the Hamming distance. However, key exceptions were found in specific classes, revealing intra-class heterogeneity. We have established that these deviations are not random but are systemic and predictable. Furthermore, we were able to quantify these irregularities, turning potential errors into manageable phenomena. The most important novelty of this work is the demarcation, for the first time to the best of our knowledge, of precise Hamming distance intervals for the classification probability. These intervals bound the possible values the probability can assume, and provide a new foundational tool for probabilistic assessment in quantum classification. Practitioners can now endorse classification results with high certainty or dismiss them with confidence. This framework can significantly enhance any quantum classification algorithm's reliability and decision-making capability.
Open AccessProbabilistic Links Between Quantum Classification of Patterns of Boolean Functions and Hamming Distance
This article investigates the probabilistic relationship between quantum classification of Boolean functions and their Hamming distance. By integrating concepts from quantum computing, information theory, and combinatorics, we explore how Hamming distance serves as a metric for analyzing deviations in function classification. Our extensive experimental results confirm that the Hamming distance is a pivotal metric for validating nearest neighbors in the process of classifying random functions. One of the significant conclusions we arrived is that the successful classification probability decreases monotonically with the Hamming distance. However, key exceptions were found in specific classes, revealing intra-class heterogeneity. We have established that these deviations are not random but are systemic and predictable. Furthermore, we were able to quantify these irregularities, turning potential errors into manageable phenomena. The most important novelty of this work is the demarcation, for the first time to the best of our knowledge, of precise Hamming distance intervals for the classification probability. These intervals bound the possible values the probability can assume, and provide a new foundational tool for probabilistic assessment in quantum classification. Practitioners can now endorse classification results with high certainty or dismiss them with confidence. This framework can significantly enhance any quantum classification algorithm's reliability and decision-making capability.
- Open Access
- 6 September 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (9) , 408
A GHZ-Based Protocol for the Dining Information Brokers Problem
This article introduces the innovative Quantum Dining Information Brokers Problem, presenting a novel entanglement-based quantum protocol to address it. The scenario involves n information brokers, all located in distinct geographical regions, engaging in a metaphorical virtual dinner. The objective is for each broker to share a unique piece of information with all the others simultaneously. Unlike previous approaches, this protocol enables a fully parallel, single-step communication exchange among all the brokers, regardless of their physical locations. A key feature of this protocol is its ability to ensure that both the anonymity and privacy of all the participants are preserved, meaning that no broker can discern the identity of the sender of any received information. At its core, the Quantum Dining Information Brokers Problem serves as a conceptual framework for achieving anonymous, untraceable, and massively parallel information exchange in a distributed system. The proposed protocol introduces three significant advancements. First, while quantum protocols for one-to-many simultaneous information transmission have been developed, this is, to the best of our knowledge, one of the first quantum protocols to facilitate many-to-many simultaneous information exchange. Second, it guarantees complete anonymity and untraceability for all senders, a critical improvement over sequential applications of one-to-many protocols, which fail to ensure such robust anonymity. Third, leveraging quantum entanglement, the protocol operates in a fully distributed manner, accommodating brokers in diverse spatial locations. This approach marks a substantial advancement in secure, scalable, and anonymous communication, with potential applications in distributed environments where privacy and parallelism are paramount.
- 6 September 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (9) , 408
Open AccessA GHZ-Based Protocol for the Dining Information Brokers Problem
This article introduces the innovative Quantum Dining Information Brokers Problem, presenting a novel entanglement-based quantum protocol to address it. The scenario involves n information brokers, all located in distinct geographical regions, engaging in a metaphorical virtual dinner. The objective is for each broker to share a unique piece of information with all the others simultaneously. Unlike previous approaches, this protocol enables a fully parallel, single-step communication exchange among all the brokers, regardless of their physical locations. A key feature of this protocol is its ability to ensure that both the anonymity and privacy of all the participants are preserved, meaning that no broker can discern the identity of the sender of any received information. At its core, the Quantum Dining Information Brokers Problem serves as a conceptual framework for achieving anonymous, untraceable, and massively parallel information exchange in a distributed system. The proposed protocol introduces three significant advancements. First, while quantum protocols for one-to-many simultaneous information transmission have been developed, this is, to the best of our knowledge, one of the first quantum protocols to facilitate many-to-many simultaneous information exchange. Second, it guarantees complete anonymity and untraceability for all senders, a critical improvement over sequential applications of one-to-many protocols, which fail to ensure such robust anonymity. Third, leveraging quantum entanglement, the protocol operates in a fully distributed manner, accommodating brokers in diverse spatial locations. This approach marks a substantial advancement in secure, scalable, and anonymous communication, with potential applications in distributed environments where privacy and parallelism are paramount.
- Open Access
Quantum Shadows: The Dining Information Brokers
This article introduces the innovative Quantum Dining Information Brokers Problem, presenting a novel entanglement-based quantum protocol to address it. The scenario involves $n$ information brokers, all located in distinct geographical regions, engaging in a metaphorical virtual dinner. The objective is for each broker to share a unique piece of information with all others simultaneously. Unlike previous approaches, this protocol enables a fully parallel, single-step communication exchange among all brokers, regardless of their physical locations. A key feature of this protocol is its ability to ensure both the anonymity and privacy of all participants are preserved, meaning no broker can discern the identity of the sender behind any received information. At its core, the Quantum Dining Information Brokers Problem serves as a conceptual framework for achieving anonymous, untraceable, and massively parallel information exchange in a distributed system. The proposed protocol introduces three significant advancements. First, while quantum protocols for one-to-many simultaneous information transmission have been developed, this is, to the best of our knowledge, one of the first quantum protocols to facilitate many-to-many simultaneous information exchange. Second, it guarantees complete anonymity and untraceability for all senders, a critical improvement over sequential applications of one-to-many protocols, which fail to ensure such robust anonymity. Third, leveraging quantum entanglement, the protocol operates in a fully distributed manner, accommodating brokers in diverse spatial locations. This approach marks a substantial advancement in secure, scalable, and anonymous communication, with potential applications in distributed environments where privacy and parallelism are paramount.Open AccessQuantum Shadows: The Dining Information Brokers
This article introduces the innovative Quantum Dining Information Brokers Problem, presenting a novel entanglement-based quantum protocol to address it. The scenario involves $n$ information brokers, all located in distinct geographical regions, engaging in a metaphorical virtual dinner. The objective is for each broker to share a unique piece of information with all others simultaneously. Unlike previous approaches, this protocol enables a fully parallel, single-step communication exchange among all brokers, regardless of their physical locations. A key feature of this protocol is its ability to ensure both the anonymity and privacy of all participants are preserved, meaning no broker can discern the identity of the sender behind any received information. At its core, the Quantum Dining Information Brokers Problem serves as a conceptual framework for achieving anonymous, untraceable, and massively parallel information exchange in a distributed system. The proposed protocol introduces three significant advancements. First, while quantum protocols for one-to-many simultaneous information transmission have been developed, this is, to the best of our knowledge, one of the first quantum protocols to facilitate many-to-many simultaneous information exchange. Second, it guarantees complete anonymity and untraceability for all senders, a critical improvement over sequential applications of one-to-many protocols, which fail to ensure such robust anonymity. Third, leveraging quantum entanglement, the protocol operates in a fully distributed manner, accommodating brokers in diverse spatial locations. This approach marks a substantial advancement in secure, scalable, and anonymous communication, with potential applications in distributed environments where privacy and parallelism are paramount.