Filters








25 Hits in 3.0 sec

TRUST: Triangle Counting Reloaded on GPUs

Santosh Pandey, Zhibin Wang, Sheng Zhong, Chen Tian, Bolong Zheng, Xiaoye Li, Lingda Li, Adolfy Hoisie, Caiwen Ding, Dong Li, Hang Liu
2021 IEEE Transactions on Parallel and Distributed Systems  
On the contrary, we advocate that i) hashing can help the key operations for scalable triangle counting on Graphics Processing Units (GPUs), i.e., list intersection and graph partitioning, ii)vertex-centric  ...  In this work, we present TRUST which performs triangle counting with the hash operation and vertex-centric mechanism at the core.  ...  CONCLUSION This paper introduces TRUST that reloads triangle counting on GPUs.  ... 
doi:10.1109/tpds.2021.3064892 fatcat:dzueeea3g5f47gty4vsbr7pkfi

Performance Analysis of Scientific Computing Workloads on Trusted Execution Environments [article]

Ayaz Akram, Anna Giannakou, Venkatesh Akella, Jason Lowe-Power, Sean Peisert
2020 arXiv   pre-print
Scientific computing sometimes involves computation on sensitive data.  ...  To study the applicability of hardware-based trusted execution environments (TEEs) to enable secure scientific computing, we deeply analyze the performance impact of AMD SEV and Intel SGX for diverse HPC  ...  Graviton [62] provides trusted execution capabilities on GPUs by providing isolation to sensitive kernels from other code executing on GPUs and the host.  ... 
arXiv:2010.13216v1 fatcat:2j25hsb3jzbgllbltmw5357eee

De-Anonymizing Text by Fingerprinting Language Generation [article]

Zhen Sun, Roei Schuster, Vitaly Shmatikov
2020 arXiv   pre-print
Secure coding practices, such as ensuring that no execution paths depend on confidential inputs, have not yet been adopted by ML developers.  ...  Related work Prior work showed how to infer model architectures and weights-but not inputs-via model execution time [9] , addresses of memory accesses leaked by GPUs [21] and trusted hardware enclaves  ...  This research will primarily benefit implementors of ML models and, in general, increase trust in ML systems. A Trace preprocessing Challenge: loop iterations are faster than Flush+Reload.  ... 
arXiv:2006.09615v2 fatcat:hwwaq2ixdfdatoncgw4ooomgpy

2021 Index IEEE Transactions on Parallel and Distributed Systems Vol. 32

2022 IEEE Transactions on Parallel and Distributed Systems  
Zhang, T., +, TPDS Feb. 2021 457-469 Arrays Trust: Triangle Counting Reloaded on GPUs.  ...  ., +, TPDS July 2021 1789-1801 Multi-GPU Design and Performance Evaluation of Homomorphic Encryption on GPU Clusters.  ...  Graph coloring Feluca: A Two-Stage Graph Coloring Algorithm With Color-Centric Paradigm on GPU. Zheng, Z., +,  ... 
doi:10.1109/tpds.2021.3107121 fatcat:e7bh2xssazdrjcpgn64mqh4hb4

CloudShield: Real-time Anomaly Detection in the Cloud [article]

Zecheng He, Ruby B. Lee
2021 arXiv   pre-print
We evaluate the proposed CloudShield on representative cloud benchmarks. Our evaluation shows that CloudShield, using model pretraining, can apply to a wide scope of cloud workloads.  ...  Specifically, the calculation of RED (t RED ) is performed on the GPU and the calculation of KDE (t KDE ) is performed on the CPU Behavior Marker Behavior Marker Behavior Marker Behavior Marker  ...  We plot three benign programs, i.e., gcc (green square), gpg (green diamond) and libquantum (green triangle) in Figure 4 .  ... 
arXiv:2108.08977v2 fatcat:37djyvh2bbcv3a3tto6lqoitbu

High performance multiscale image processing framework on multi-gpus (graphics processing units) with applications to unbiased diffeomorphic atlas construction

Linh Khanh Ha
2012
the GPUs.  ...  for scientists to experiment and evaluate the influence of these techniques on their research and practical applications.  ...  Parallel computing now has a big influence on everyday life and research by providing more accurate, detailed, and trusted predictions.  ... 
doi:10.26053/0h-2zzt-x9g0 fatcat:aih5hfxbqvef7jrlp533ply3uu

Computer-Assisted Interactive Documentary and Performance Arts in Illimitable Space [article]

Miao Song
2012 arXiv   pre-print
It focuses on the advanced computer graphics and emerges from experimental cinematic works and theatrical artistic practices.  ...  It also includes building up on this experience to integrate Kinect- and haptic-based interaction, artistic scenery rendering, and other forms of control.  ...  In order to achieve social realism, artists not only rely on other visual, character, motion realism, but also count on the writing of the documentary production.  ... 
arXiv:1212.6250v1 fatcat:2bulqbnpsbc3daw5s6xeqvk26y

Towards exascale computing for cosmological simulations

James Douglas Potter
2017
I'm sure there is even more in store before we are finally forced to retire and work on building our respective wine cellars.  ...  He also introduced me to the science fair which had a positive impact on my education. By far the greatest influence has been my friend Joachim Stadel.  ...  On each node of Piz Daint there are 8 CPU-cores and one Nvidia Tesla K20X GPU accelerator.  ... 
doi:10.5167/uzh-204887 fatcat:xg5dng5pw5enpirysejlqexq2i

Altered Perceptions Discerning the Natural from the Virtual in Extended Reality [article]

Neil Christensen, University Of Calgary, Christian Jacob, Gerald Hushlak
2019
Upon closer inspection and study, the fault of low polygon counts required by XR development also proved to be problematic as triangles were easily detected on the texture map projection when viewed in  ...  CPU plus GPU combined) approach reduced the possibilities as visual post-processing effects, resolution per eye and polygon counts for 3D models were greatly limited.  ... 
doi:10.11575/prism/37321 fatcat:ctbwpadz2jdezetsr6p2aam37y

Heiko Falk OASIcs-OpenAccess Series in Informatics

Heiko Falk, Daniel Cremers, Barbara Hammer, Marc Langheinrich, Dorothea Wagner
2014 14th International Workshop on Worst-Case Execution Time Analysis WCET 2014   unpublished
This work was partially supported by COST Action IC1202: Timing Analysis On Code-Level (TACLe), and by the Swedish Research Council project Contesse (2010-4276). Acknowledgement.  ...  Only the upper right triangle of the matrix gets accessed.  ...  To further improve trust, such scenarios could be verified, whenever possible, by automated tools.  ... 
fatcat:ncfh6om6q5hjpdtnmqingr4v4u

Spatiotemporal enabled Content-based Image Retrieval

Mariana Belgiu, Martin Sudmanns, Tiede Dirk, Andrea Baraldi, Stefan Lang
2016 International Conference on GIScience Short Paper Proceedings  
The efficiency of the coverage of a sensor network depends on optimal position of each sensor node within the network.  ...  A sensor in WSNs covers only a limited region, depending on its sensing and communicating ranges, as well as the environment configuration.  ...  Demšar who read and commented on an earlier version.  ... 
doi:10.21433/b311729295dw fatcat:fulw4pw3kfh5nmfzcsy3pkisvm

Pencil Code [article]

on behalf of the Pencil Code Collaboration Axel Brandenburg
2020 Zenodo  
In the beginning of each sub-time-step particles are counted in blocks and the block counts communicated to the bricks on the parent processors.  ...  Particles on each processor are counted in bricks of size nbx×nby×nbz (typically nbx= nby=nbz=4). 3.  ...  . . . . . . . . . . . . . . . . . . . 204 gmu5mz . . . . . . . . . . . . . . . . . . . . . 204 gmu5rms . . . . . . . . . . . . . . . . . . . . 204 gmuSrms . . . . . . . . . . . . . . . . . . . . 204 gpu  ... 
doi:10.5281/zenodo.3961647 fatcat:wrpaz23op5a2doysqq3zk7xme4

The Pencil Code, a modular MPI code for partial differential equations and particles: multipurpose and multiuser-maintained [article]

on behalf of the Pencil Code Collaboration Axel Brandenburg
2021 Zenodo  
In the beginning of each sub-time-step particles are counted in blocks and the block counts communicated to the bricks on the parent processors.  ...  Particles on each processor are counted in bricks of size nbx×nby×nbz (typically nbx= nby=nbz=4). 3.  ...  . . . . . . . . . . . . . . . . . . . 197 gmu5mz . . . . . . . . . . . . . . . . . . . . . 197 gmu5rms . . . . . . . . . . . . . . . . . . . . 197 gmuSrms . . . . . . . . . . . . . . . . . . . . 197 gpu  ... 
doi:10.5281/zenodo.4553325 fatcat:2icmms6n3zawfmoceifloeo6me

Effects and opportunities of native code extensions for computationally demanding web applications [article]

Dennis Jarosch, Humboldt-Universität Zu Berlin, Humboldt-Universität Zu Berlin
2017
On the other hand, it shows substantial limitations in data processing. These are evaluated and possible solutions are discussed.  ...  The results of the performance analysis are complemented with an evaluation on the basis of technical and non-technical criteria and a discussion of the technical, political, and strategic drivers for  ...  The OpenGL API supports three different triangle primitives. GL_TRIANGLES draws separate triangles with exactly three vertices per triangle.  ... 
doi:10.18452/16451 fatcat:racqnzleubghfkj4doyzueqmqu

Overcoming the Intuition Wall: Measurement and Analysis in Computer Architecture

John David Demme
2017
No one architect can now understand all the complexities of many systems and reason about the full impact of changes or new applications.  ...  In each case study we find at least one surprising or unintuitive result which would likely not have been found without the application of our method.  ...  The reason for redundancy follows from the triangle inequality property of distance computations.  ... 
doi:10.7916/d8x0652n fatcat:ypxoqhjfuff7deexy5uxrlryze
« Previous Showing results 1 — 15 out of 25 results