Efficient Neighbor-Sampling-based GNN Training on CPU-FPGA Heterogeneous Platform

Zhang, Bingyi; Kuppannagari, Sanmukh R.; Kannan, Rajgopal; Prasanna, Viktor K.

doi:10.1109/hpec49654.2021.9622822

Cited by 10 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CPU-FPGA heterogeneous approaches, in particular, offer solutions for applications requiring high levels of parallelism and customizability. By combining the general-purpose processing capability of the CPU with the specialized hardware advantages of the FPGA, these approaches enable the faster and more energy-efficient performance of complex computational tasks [27]. Although these structures are not applicable to embedded systems, they are preferred for large-scale applications.…”

Section: Fpga-based Heterogeneous Approachesmentioning

confidence: 99%

“…Zhang et al [27] recommend a method for training GNNs on large-scale graphs using a CPU-FPGA heterogeneous platform. The method uses neighbor sampling to address scalability and overfitting challenges.…”

Section: Fpga-based Heterogeneous Approachesmentioning

confidence: 99%

“…While these studies cover both traditional and embedded hardware, GPU and CPU implementations form a large part of the overall topic. Although the use of low-bit precision and FPGA-based accelerators to derive low-dimensional and real-time models is a well-known research topic in Neural Networks (NNs) [25] and Convolutional Neural Networks [26], there has been insufficient research on GNNs [20,24,27,28]. Therefore, the study of quantized GNN models for accelerators is a promising potential research topic.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Kose,

Nunez-Yanez,

Piechocki

et al. 2024

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) are powerful models capable of managing intricate connections in non-Euclidean data, such as social networks, physical systems, chemical structures, and communication networks. Despite their effectiveness, the large-scale and complex nature of graph data demands substantial computational resources and high performance during both training and inference stages, presenting significant challenges, particularly in the context of embedded systems. Recent studies on GNNs have investigated both software and hardware solutions to enhance computational efficiency. Earlier studies on Deep Neural Networks (DNNs) have indicated that methods like reconfigurable hardware and quantization are beneficial in addressing these issues. Unlike DNN research, studies on efficient computational methods for GNNs are less developed and require more exploration. This survey reviews the latest developments in quantization and FPGA-based acceleration for GNNs, showcasing the capabilities of reconfigurable systems (often FPGAs) to offer customized solutions in environments marked by significant sparsity and the necessity for dynamic load management. It also emphasizes the role of quantization in reducing both computational and memory demands through the use of fixed-point arithmetic and streamlined vector formats. This paper concentrates on low-power, resource-limited devices over general hardware accelerators and reviews research applicable to embedded systems. Additionally, it provides a detailed discussion of potential research gaps, foundational knowledge, obstacles, and prospective future directions.

show abstract

Section: Fpga-based Heterogeneous Approachesmentioning

confidence: 99%

Section: Fpga-based Heterogeneous Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Kose,

Nunez-Yanez,

Piechocki

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Paul et al 33 constructed the capsule attention‐based mechanism model based on GNN for job allocation in a heterogeneous team with deadline constraints. The GNN has the over‐smoothing and neighborhood explosion challenges which are addressed in recent studies 40,41 …”

Section: Introductionmentioning

confidence: 99%

Distributed job allocation using response threshold for heterogeneous robot team under deadline constraints

Vivek¹,

Kumari²

2023

Concurrency and Computation

View full text Add to dashboard Cite

Summary Job allocation among the robots is a significant challenge in a multi‐robot environment. This article proposes a distributed algorithm for job allocation with deadline constraints. It works utilizing the response threshold model inspired by job distribution among insect societies in nature. The team divides the jobs among themselves based on their threshold capacity. The proposed approach focus on maximizing the total jobs completed with the available resources. The static case where the robots never fail is considered first for evaluating the algorithm. Then it is run in a dynamic environment with robot failure to study the effects on job allocation. Two experimental studies are conducted with varying amounts of resources. In static and dynamic cases under experimental study I, the jobs completion percentage increased by an average of 37.6% and 35.8%, respectively, over the existing method. Under experimental study II, the proposed algorithm completed all the jobs with improved resource utilization. The results show that the proposed algorithm achieves better jobs completion. The algorithm is then tested in a real‐world experimental environment using FireBird V mobile robots under static and dynamic cases. The outcomes indicate that the proposed approach is efficient in job allocation in practical circumstances.

show abstract

Real-Time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics

Neu,

Becker,

Dorwarth

et al. 2024

Comput Softw Big Sci

View full text Add to dashboard Cite

We present a design methodology that enables the semi-automatic generation of a hardware-accelerated graph building architectures for locally constrained graphs based on formally described detector definitions. In addition, we define a similarity measure in order to compare our locally constrained graph building approaches with commonly used k-nearest neighbour building approaches. To demonstrate the feasibility of our solution for particle physics applications, we implemented a real-time graph building approach in a case study for the Belle II central drift chamber using Field-Programmable Gate Arrays (FPGAs). Our presented solution adheres to all throughput and latency constraints currently present in the hardware-based trigger of the Belle II experiment. We achieve constant time complexity at the expense of linear space complexity and thus prove that our automated methodology generates online graph building designs suitable for a wide range of particle physics applications. By enabling an hardware-accelerated preprocessing of graphs, we enable the deployment of novel Graph Neural Networks (GNNs) in first-level triggers of particle physics experiments.

show abstract

Efficient Neighbor-Sampling-based GNN Training on CPU-FPGA Heterogeneous Platform

Cited by 10 publications

References 21 publications

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

A Survey of Computationally Efficient Graph Neural Networks for Reconfigurable Systems

Distributed job allocation using response threshold for heterogeneous robot team under deadline constraints

Real-Time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics

Contact Info

Product

Resources

About