WolfGraph: The edge-centric graph processing on GPU

Zhu, Huanzhou; He, Ligang; Leeke, Matthew; Mao, Rui

doi:10.1016/j.future.2019.09.052

Cited by 10 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first limitation, i.e., the computational time, can be addressed by further fine-tuning the system's hyper-parameters (e.g., the number of NGR iterations, the convergence window size, the convergence threshold) without compromising the performance, so to set the optimal trade-off between computational efficiency and the detection performance as per the clinical needs. Apart from this, dedicated hardware accelerators for graph processing can also be envisaged further to improve the execution time of the proposed scheme [38]. The second limitation of the proposed scheme is its dependency on computing the nuclei centroid within the WSI patches using the detection backbone.…”

Section: F Limitationsmentioning

confidence: 99%

Neural Graph Refinement for Robust Recognition of Nuclei Communities in Histopathological Landscape

Hassan,

Li,

Javed

et al. 2024

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Accurate classification of nuclei communities is an important step towards timely treating the cancer spread. Graph theory provides an elegant way to represent and analyze nuclei communities within the histopathological landscape in order to perform tissue phenotyping and tumor profiling tasks. Many researchers have worked on recognizing nuclei regions within the histology images in order to grade cancerous progression. However, due to the high structural similarities between nuclei communities, defining a model that can accurately differentiate between nuclei pathological patterns still needs to be solved. To surmount this challenge, we present a novel approach, dubbed neural graph refinement, that enhances the capabilities of existing models to perform nuclei recognition tasks by employing graph representational learning and broadcasting processes. Based on the physical interaction of the nuclei, we first construct a fully connected graph in which nodes represent nuclei and adjacent nodes are connected to each other via an undirected edge. For each edge and node pair, appearance and geometric features are computed and are then utilized for generating the neural graph embeddings. These embeddings are used for diffusing contextual information to the neighboring nodes, all along a path traversing the whole graph to infer global information over an entire nuclei network and predict pathologically meaningful communities. Through rigorous evaluation of the proposed scheme across four public datasets, we showcase that learning such communities through neural graph refinement produces better results that outperform state-of-the-art methods.

show abstract

Section: F Limitationsmentioning

confidence: 99%

Neural Graph Refinement for Robust Recognition of Nuclei Communities in Histopathological Landscape

Hassan,

Li,

Javed

et al. 2024

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…By transposing the image in Figure 5, adjacent threads generate adjacent addresses. As there are now regular access patterns unlike interleaved, many accesses otherwise are coalesced to one, improving memory bandwidth utilization to achieve peak performance [25]. Figure 4 shows the coalesced and non-coalesced accesses.…”

Section: Accelerated Logistic Map Image Cryptosystemsmentioning

confidence: 99%

An Accelerator-based Logistic Map Image Cryptosystems for Grayscale Images

Holla

Pais

Suma

2021

JCSANDM

View full text Add to dashboard Cite

The logistic map is a class of chaotic maps. It is still in use in image cryptography. The logistic map cryptosystem has two stages, namely permutation, and diffusion. These two stages being computationally intensive, the permutation relocates the pixels, whereas the diffusion rescales them. The research on refining the logistic map is progressing to make the encryption more secure. Now there is a need to improve its efficiency to enable such models to fit for high-speed applications. The new invention of accelerators offers efficiency. But the inherent data dependencies hinder the use of accelerators. This paper discusses the novelty of identifying independent data-parallel tasks in a logistic map, handing them over to the accelerators, and improving their efficiency. Among the two accelerator models proposed, the first one achieves peak efficiency using coalesced memory access. The other cryptosystem further improves performance at the cost of more execution resources. In this investigation, it is noteworthy that the parallelly accelerated logistic map achieved a significant speedup to the larger grayscale image used. The objective security estimates proved that the two stages of the proposed systems progressively ensure security.

show abstract

“…Several general-purpose graph processing frameworks have been introduced and evaluated in recent years to improve the programmability of graph applications, mainly focusing on performance and/or scalability. These frameworks include, but not limited to, vertex-centric [10,21,26,27], edge-centric [42,56,57], data-centric [36], matrix based [7,50], task based [23] and declarative programming based [45]. Among all these models, the idea of "think like a vertex" or vertex-centric programming model has seen significant interest and widespread deployment in recent works [15,24,26], and is our model of choice as well.…”

Section: Think Like a Vertexmentioning

confidence: 99%

GraphGuess: Approximate Graph Processing System with Adaptive Correction

Ramezani¹,

Kandemir²,

Sivasubramaniam³

2021

Preprint

View full text Add to dashboard Cite

Graph-based data structures have drawn great attention in recent years. The large and rapidly growing trend on developing graph processing systems focus mostly on improving the performance by preprocessing the input graph and modifying its layout. These systems usually take several hours to days to complete processing a single graph on high-end machines, let alone the overhead of preprocessing which most of the time can be dominant. Yet for most of graph applications the exact answer is not always crucial, and providing a rough estimate of the final result is adequate. Approximate computing is introduced to trade off accuracy of results for computation or energy savings that could not be achieved by conventional techniques alone. Although various computing platforms and application domains benefit from approximate computing, it has not been thoroughly explored yet in the context of graph processing systems.In this work, we design, implement and evaluate GraphGuess, inspired from the domain of approximate graph theory and extend it to a general, practical graph processing system. GraphGuess is essentially an approximate graph processing technique with adaptive correction, which can be implemented on top of any graph processing system. We build a vertex-centric processing system based on GraphGuess, where it allows the user to trade off accuracy for better performance. Our experimental studies show that using GraphGuess can significantly reduce the processing time for large scale graphs while maintaining high accuracy.

show abstract

WolfGraph: The edge-centric graph processing on GPU

Abstract: Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

Cited by 10 publications

References 33 publications

Neural Graph Refinement for Robust Recognition of Nuclei Communities in Histopathological Landscape

Neural Graph Refinement for Robust Recognition of Nuclei Communities in Histopathological Landscape

An Accelerator-based Logistic Map Image Cryptosystems for Grayscale Images

GraphGuess: Approximate Graph Processing System with Adaptive Correction

Contact Info

Product

Resources

About