Mosaic: Processing a trillion-edge graph on a single machine

S Maass, C Min, S Kashyap, W Kang… - Proceedings of the …, 2017 - dl.acm.org
Processing a one trillion-edge graph has recently been demonstrated by distributed graph
engines running on clusters of tens to hundreds of nodes. In this paper, we employ a single …

Gluon: A communication-optimizing substrate for distributed heterogeneous graph analytics

R Dathathri, G Gill, L Hoang, HV Dang… - Proceedings of the 39th …, 2018 - dl.acm.org
This paper introduces a new approach to building distributed-memory graph analytics
systems that exploits heterogeneity in processor types (CPU and GPU), partitioning policies …

Mariusgnn: Resource-efficient out-of-core training of graph neural networks

R Waleffe, J Mohoney, T Rekatsinas… - Proceedings of the …, 2023 - dl.acm.org
We study training of Graph Neural Networks (GNNs) for large-scale graphs. We revisit the
premise of using distributed training for billion-scale graphs and show that for graphs that fit …

Automatic cricket highlight generation using event-driven and excitement-based features

P Shukla, H Sadana, A Bansal… - Proceedings of the …, 2018 - openaccess.thecvf.com
Producing sports highlights is a labor-intensive work that requires some degree of
specialization. We propose a model capable of automatically generating sports highlights …

TurboGraph++ A scalable and fast graph analytics system

S Ko, WS Han - Proceedings of the 2018 international conference on …, 2018 - dl.acm.org
Existing distributed graph analytics systems are categorized into two main groups: those that
focus on efficiency with a risk of out-of-memory error and those that focus on scale-up with a …

The computational sprinting game

S Fan, SM Zahedi, BC Lee - ACM SIGARCH Computer Architecture …, 2016 - dl.acm.org
Computational sprinting is a class of mechanisms that boost performance but dissipate
additional power. We describe a sprinting architecture in which many, independent chip …

Parallel strong connectivity based on faster reachability

L Wang, X Dong, Y Gu, Y Sun - Proceedings of the ACM on Management …, 2023 - dl.acm.org
Computing strongly connected components (SCC) is among the most fundamental problems
in graph analytics. Given the large size of today's real-world graphs, parallel SCC …

Speeding up SpMV for power-law graph analytics by enhancing locality & vectorization

S Yesil, A Heidarshenas, A Morrison… - … Conference for High …, 2020 - ieeexplore.ieee.org
Graph analytics applications often target large-scale web and social networks, which are
typically power-law graphs. Graph algorithms can often be recast as generalized Sparse …

Gluon-async: A bulk-asynchronous system for distributed and heterogeneous graph analytics

R Dathathri, G Gill, L Hoang, V Jatala… - 2019 28th …, 2019 - ieeexplore.ieee.org
Distributed graph analytics systems for CPUs, like D-Galois and Gemini, and for GPUs, like
D-IrGL and Lux, use a bulk-synchronous parallel (BSP) programming and execution model …

High-Performance and Flexible Parallel Algorithms for Semisort and Related Problems

X Dong, Y Wu, Z Wang, L Dhulipala, Y Gu… - Proceedings of the 35th …, 2023 - dl.acm.org
Semisort is a fundamental algorithmic primitive widely used in the design and analysis of
efficient parallel algorithms. It takes input as an array of records and a function extracting a …