We study exascale parallel algorithms for the selection of intervention or monitoring strategies in massive realistic socio-technical networks through scalable Influence Maximization (InfMax) algorithms.We employ novel techniques to enable efficient scaling on up to 8k nodes of OLCF Frontier, with 65k AMD GPUs and 458k AMD CPU cores.Current state-of-the-art InfMax tools are limited to networks with only a few million actors (vertices) and a few hundred million interactions (edges).By overcoming these limitations, ICS '25, June 08-11, 2025, Salt Lake City, UT, USA Minutoli et al.we show that our approach is capable of processing a realistic social contact network of the United States with 285 million nodes and about 8 billion edges.This two ordersof-magnitude improvement over the previous state-of-theart is obtained by leveraging algorithmic advancements for the InfMax problem and designing several problem-specific approaches to overlap communication with computation, improve GPU efficiency, and lower the application's memory requirements.We evaluate strong scaling for computing 10k most influential seeds using up to 8k nodes of an exascale system, and weak scaling from 128 to 8k system nodes for seed sets ranging from 625 to 40k seeds.We achieve the fastest-known runtime of 25 minutes while performing 48 million diffusion simulations totaling 2.31 petabytes to identify 40k influential seeds using 8k nodes, and take 5.75 minutes to identify 10k seeds while using 4k nodes.

The rapid growth of large-scale datasets in fields like biology and social networks has driven the need for advanced graph analytics techniques. Community detection, a fundamental task in graph analytics, identifies closely connected groups of nodes within a network, providing valuable insights across various disciplines. This study focuses on two classic community detection methods, the Louvain algorithm and Markov Clustering (MCL), and evaluates the performance of two prominent distributed community detection algorithms: HiPDPL-GPU, our prior implementation, and HipMCL. We conduct experiments on GPU-accelerated heterogeneous HPC systems, Summit and Frontier, to assess their performance under varying conditions. Our objective is to identify the strengths and weaknesses of these algorithms in terms of scalability, and quality of solutions. We evaluate these algorithms on a diverse set of 70+ networks spanning 13 domains, with sizes ranging up to 4.2 billion edges. Our results demonstrate that HiPDPL-GPU consistently outperforms HipMCL, especially for large-scale networks. HiPDPL-GPU achieves significantly faster runtimes (47x to 1439x), higher modularity scores, and improved scalability. These findings highlight HiPDPL-GPU as a promising solution for efficient and effective large-scale graph analytics in diverse application domains, and provide insights into the feasibility of using MCL-based approaches for certain application domains.

The power & energy demands of HPC machines have grown significantly. Modern exascale HPC systems require tens of megawatts of combined power for computing resources and cooling facilities at full capacity. The current energy trend is not sustainable for future HPC systems, and there is a need to work toward the energy efficiency aspect of HPC performance. Energy awareness of the HPC applications at the job level is essential for running an efficient HPC system. This work aims to develop a pipeline to provide a production-level system-wide overview of the HPC workloads' power profile while handling evolving workloads exhibiting new power trends. We developed an open-set classification model for HPC jobs based on the properties of power profiles to continuously provide a system-wide holistic view of recently completed jobs. The pipeline helps continuously monitor the job-level power usage pattern of HPC and enables us to capture the new trends in applications' power behavior. We employed a comprehensive set of techniques to generate job-level data, custom-designed feature extraction methods to extract critical features from jobs' power profiles, clustering techniques powered by generative modeling, and open-set classification for identifying job profiles into known classes or an unknown set. With extensive evaluations, we demonstrate the effectiveness of each component in our pipeline. We provide an analysis of the resulting clusters that characterize the power profile landscape of the Summit supercomputer from more than 60K jobs executed in a year. The open-set classification classifies the known data sets into known classes with high accuracy and identifies unknown data noints with over 85% accuracy.

Graph Convolutional Network (GCN) has drawn considerable research attention in recent times. Many different problems from diverse domains can be solved efficiently using GCN. Community detection in graphs is a computationally challenging graph analytic problem. The presence of only a limited amount of labelled data (known communities) motivates us for using a learning approach to community discovery. However, detecting communities in large graphs using semi-supervised learning with GCN is still an open problem due to the scalability and accuracy issues. In this paper, we present a scalable method for detecting communities based on GCN via semi-supervised node classification. We optimize the hyper-parameters for our semi-supervised model for detecting communities using PyTorch with CUDA on GPU environment. We apply Mini-batch Gradient Descent for larger datasets to resolve the memory issue. We demonstrate an experimental evaluation on different real-world networks from diverse domains. Our model achieves up to 86.9% accuracy and 0.85 F1 Score on these practical datasets. We also show that using identity matrix as features, based on the graph connectivity, performs better with higher accuracy than that of vertex-based graph features. We accelerate the model performance 4 times with the use of GPUs over CPUs.

Web spam is a serious threat for both end-users and search engines (w.r.t., query cost). Webgraphs can be exploited in detecting spam. In the past, several graph mining techniques were applied to measure metrics for pages and hyperlinks. In this paper, we justify the importance of webgraph to distinguish spam websites from non-spam ones based on several graph metrics computed for a labelled dataset (WEBSPAM-UK2007) and justify our model by testing on uk-2014 dataset, the most recently available dataset on the same (uk) domain. WEBSPAM-UK2007 dataset includes 0.1 million different hosts and four kinds of feature sets: Obvious, Link, Transformed Link and Content. We use five prominent machine learning (ML) techniques (i.e., Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Logistic Regression, Naïve Bayes and Random Forest) to build a ML-based classifier. To evaluate the performance of our classifier, we compute accuracy and F-1 score and perform 10-fold cross validation. We also compare graph based features with content based textual features and find that graph properties are similar or better than text properties. We achieve above 99% training accuracy for most of our machine learning models. We test our model with uk-2014 dataset with 4.7 million hosts for the graph-based feature sets and achieve accuracy in between 90-94% for most of the models. To the best of our knowledge, prior works on web spam detection with WEBSPAM-UK2007 dataset did not use different test dataset for their models. Our model classifier is capable of detecting web spam for any input webgraph based on its graph metrics features.