14 Hits in 4.7 sec

Enhancing Programmability, Portability, and Performance with Rich Cross-Layer Abstractions [article]

Nandita Vijaykumar
2019 arXiv   pre-print
These abstractions are designed to communicate higher-level program information from the application to the underlying system and hardware in a highly efficient manner, requiring only minor additions to  ...  designing richer abstractions between the application, system software, and hardware architecture in different contexts to significantly improve programmability, portability, and performance in CPUs and GPUs  ...  (ii) Zorua can be utilized for low-latency preemption of GPU applications, by leveraging the ability to swap in/out resources from/to memory in a transparent manner.  ... 
arXiv:1911.05660v1 fatcat:w5f3g4isqbcphm2jjfzjtvrjnq

Dataflow-Based Design and Implementation of Image Processing Applications [chapter]

Chung-Ching Shen, William Plishker, Shuvra Bhattacharyya
2012 Multimedia Image and Video Processing, Second Edition  
To demonstrate dataflow-based design methods in a manner that is concrete and easily adapted to different platforms and back-end design tools, we present in this report a number of case studies based on  ...  Dataflow is a well known computational model and is widely used for expressing the functionality of digital signal processing (DSP) applications, such as audio and video data stream processing, digital  ...  Background on GPUs for Image Processing Graphics processing units (GPUs) provide another class of high performance computing platforms that can be leveraged for image processing.  ... 
doi:10.1201/b11716-31 fatcat:wqlvs5ue75hefpiu3ftxa3yvbq

GIS Mashups [chapter]

Ilya Zaslavsky
2017 Encyclopedia of GIS  
It took more than a decade from this point for the larger computer science community to investigate GPs for pattern analysis purposes.  ...  ., modeling a configuration space for crystallographic design, casting folding energies as a function of a protein's contact map, and formulation of vaccination policies taking into account social dynamics  ...  and computation; (6) Copy output data from GPU to CPU; (7) Free the allocated GPU memory.  ... 
doi:10.1007/978-3-319-17885-1_530 fatcat:rrr5buo3zrevpigdtjwhewipvm

Microwave Radiometer RFI Detection Using Deep Learning

Priscilla N Mohammed, Jeffrey R. Piepmeier
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Radio frequency interference (RFI) is a risk for microwave radiometers due to their required very high sensitivity.  ...  This paper presents a deep learning approach to RFI detection using SMAP spectrogram data as input images.  ...  With the advent of graphics processing unit (GPU) technology for space use and the growing number of channels in digital receivers, it may be advantageous to run a deep learning algorithm on a GPU rather  ... 
doi:10.1109/jstars.2021.3091873 fatcat:inokfs4eo5fpzefrwkjl7qevve

Facilitating Typhoon-Triggered Flood Disaster-Ready Information Delivery Using SDI Services Approach—A Case Study in Hainan

Lei Hu, Zhe Fang, Mingda Zhang, Liangcun Jiang, Peng Yue
2022 Remote Sensing  
meet the specific needs of many regional stakeholders worldwide; traditional approaches with field surveys are labor-intensive, time-consuming, and expensive, especially for severe disasters that affect a  ...  In order to tackle some of the above challenges, this paper demonstrates how to facilitate typhoon-triggered flood disaster-ready information delivery using an SDI services approach, and proposes a web-based  ...  There are also many studies leveraging open standards and SDI to support a wide range of disasters, such as floods, drought, landslides, and public health.  ... 
doi:10.3390/rs14081832 fatcat:cckjsphgyjci7najviiws6iqyi

Energy-Oriented Partial Desktop Virtual Machine Migration

Nilton Bila, Eric J. Wright, Eyal De Lara, Kaustubh Joshi, H. Andrés Lagar-Cavilla, Eunbyung Park, Ashvin Goel, Matti Hiltunen, Mahadev Satyanarayanan
2015 ACM Transactions on Computer Systems  
However, desktop VMs are often large, requiring gigabytes of memory. Consolidating such VMs creates large network transfers lasting in the order of minutes and utilizes server memory inefficiently.  ...  It creates a partial replica of the desktop VM on the consolidation server by copying only VM metadata, and it transfers pages to the server on-demand, as the VM accesses them.  ...  Table III presents the desktop's power profile obtained with a GW Instek GPM-8212 power meter.  ... 
doi:10.1145/2699683 fatcat:upw4p4dekzbqxcjm4d54kjxzje

How can big data and machine learning benefit environment and water management: A survey of methods, applications, and future directions

Alexander Y. Sun, Bridget R Scanlon
2019 Environmental Research Letters  
Acknowledgments The authors were partially supported by funding from Jackson School of Geosciences, the University of Texas at Austin.  ...  In general, large-scale learning can get help from scalable parallel solvers, the efficient use of in-memory processing to reduce data transfer cost, and hardware acceleration through GPUs or Field Programmable  ...  Unlike MapReduce which persists interim datasets to local disks, SPARK performs in-memory processing of data and can be up to 100 times faster than MapReduce [56] .  ... 
doi:10.1088/1748-9326/ab1b7d fatcat:vx4thuy45vhlnmhu7bk2hwh2g4

Algorithm-Architecture Co-Design for Domain-Specific Accelerators in Communication and Artificial Intelligence

Yaoyu Tao, University, My
We present HiMA, a tiled, history-based memory access [...]  ...  This work advances LDPC codes in two aspects: 1) we take the inspiration from simulated annealing to generalize the post-processor design using three methods: quenching, extended heating, and focused heating  ...  It has been an honor to do research under his supervision from my undergraduate years and later started doing Ph.D with him. His tremendous effort, valuable ideas, and warm  ... 
doi:10.7302/4622 fatcat:bicba3kdd5e5pe6zq4psloamne

Google Earth Engine Cloud Computing Platform for Remote Sensing Big Data Applications: A Comprehensive Review

Meisam Amani, Arsalan Ghorbanian, Seyed Ali Ahmadi, Mohammad Kakooei, Armin Moghimi, S. Mohammad Mirmazloumi, Sayyed Hamed Alizadeh Moghaddam, Sahel Mahdavi, Masoud Ghahremanloo, Saeid Parsian, Qiusheng Wu, Brian Brisco
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
It was generally observed that the number of GEE publications has significantly increased during the past few years, and it is expected that GEE will be utilized by more users from different fields to  ...  In this regard, Google has developed a cloud computing platform, called Google Earth Engine (GEE), to effectively address the challenges of big data analysis.  ...  Therefore, persistent and precise monitoring of all types of water resources is a vital need.  ... 
doi:10.1109/jstars.2020.3021052 fatcat:pudllv5h2ve4lfvqtx3p4qikju

Machine Learning Approaches to Extreme Weather Events Forecast in Urban Areas: Challenges and Initial Results

2022 Supercomputing Frontiers and Innovations  
the dynamics of the phenomenon, which challenges numerical weather prediction models and opens an opportunity for Machine Learning (ML) based models that may learn complex mappings between input-output from  ...  The authors also thank professor Pedro Dias and professor Demerval Moreira, both from IAG, for providing us with NWP simulation data used in our experiments.  ...  Additionally, we use a subset of the rainfall dataset from NASA's TRMM and GPM missions, with rainfall collected for the same spatial region as in the CSFR dataset over 22 years (rainfall dataset) [14  ... 
doi:10.14529/jsfi220104 fatcat:hwdwptwbsbd4xfgz7e3sodixuq

Computational Solutions For Quality Control Of Mass Spectrometry-Based Proteomics

Wout Bittremieux, Kris Laukens, Bart Goethals
2017 Zenodo  
Third, we present an unsupervised outlier detection workflow to automatically discriminate low-quality mass spectrometry experiments from high-quality mass spectrometry experiments.  ...  defined; (ii) the basic technical infrastructure to unambiguously store and communicate quality control data has to be available; (iii) advanced analysis techniques are needed to derive actionable insights from  ...  The common Repository of Adventitious Proteins (cRAP) [53] provides a resource of contaminant proteins, sourced from the Global Proteome Machine (GPM) [60] .  ... 
doi:10.5281/zenodo.1059122 fatcat:qt2t2z3kkbg4zaixzndovabkcq

Improving the accuracy of weed species detection for robotic weed control in complex real-time environments

Alex Olsen
Once the image data from all cameras are ready, prepare the data into a batch allocated to CUDA memory on the Jetson TX2 GPU. • Perform inference on the CUDA batch and extract the resulting score and classification  ...  Intel R Core TM i7 Processor and without a high-end GPU.  ... 
doi:10.25903/vhbh-w150 fatcat:ckeopnficbfftofpslwvxtsiqq

LBNO-DEMO: Large-scale neutrino detector demonstrators for phased performance assessment in view of a long-baseline oscillation experiment [article]

L. Agostino, D. Autiero , A. Blondel, A. Bravar, M. Calin, D. Chesneanu , J. Dawson, A. Delbart , I. Efthymiopoulos, T. Esanu , A. Gendotti, P. Gorodetzky, M. Ieva, A. Korzenev, I. Lazanu, R.M. Margineanu (+67 others)
2014 arXiv   pre-print
For the near detector, a high-pressure gas TPC embedded in a calorimeter and a magnet is the baseline design.  ...  LBNO considers three types of neutrino detector technologies: a double-phase liquid argon (LAr) TPC and a magnetised iron detector as far detectors.  ...  Acknowledgements We are also grateful to the CERN Management for their encouragements, recognising the importance of CERN in the coherent definition of a potential future program for the European neutrino  ... 
arXiv:1409.4405v1 fatcat:yrurslistbctvmhstuojof4mpm

Leveraging Artificial Neural Networks for Modeling Hydrogeological Time Series

Andreas Wunsch, Nico Goldscheider, Anne Johannet
Untersucht werden hierbei Nonlinear Autoregressive Models with Exogenous Inputs (NARX), Long Short-Term Memory Networks [...]  ...  Using a GPU instead of a CPU is not possible for NARX models in our case because of the Levenberg-Marquardt training algorithm, which is not suitable for GPU computation.  ...  However, both LSTMs and CNNs can be calculated on a GPU, which in the case of LSTMs is the preferred option.  ... 
doi:10.5445/ir/1000149192 fatcat:h3gpwpsk4ngefdotygo7s6oacy