Filters








613 Hits in 2.7 sec

Accelerating Cosmological Data Analysis with FPGAs

Volodymyr V. Kindratenko, Robert J. Brunner
2009 2009 17th IEEE Symposium on Field Programmable Custom Computing Machines  
Example Analysis: Angular Correlation • Two-point angular correlation function (TPACF), (), describes the frequency distribution of angular separations  between celestial objects in the interval (,  ...  ); computeDD(data, npd, data, npd, 1,binb, nbins, njks, DD); for (i = 0; i < random_count; i++) // loop through random data files { loadRandomData(random[i]); computeRR(random[i], npr[i], random[  ... 
doi:10.1109/fccm.2009.12 dblp:conf/fccm/KindratenkoB09 fatcat:5ve47dahjzbjrcckekplvxs4bi

Accelerate Scientific Deep Learning Models on Heterogeneous Computing Platform with FPGA

Chao Jiang, David Ojika, Sofia Vallecorsa, Thorsten Kurth, Prabhat, Bhavesh Patel, Herman Lam, C. Doglioni, D. Kim, G.A. Stewart, L. Silvestris, P. Jackson (+1 others)
2020 EPJ Web of Conferences  
Heterogeneous computing (HGC), with CPUs integrated with GPUs, FPGAs, and other science-targeted accelerators, offers unique capabilities to accelerate DNNs.  ...  ) equipped with an Arria 10 GX FPGA.  ...  The training data is based on pseudo-data simulated with GEANT4 [22] in the proposed Linear Collider Detector (LCD) for the CLIC accelerator [23] .  ... 
doi:10.1051/epjconf/202024509014 fatcat:sqb2cqag6ng2njpgtjouhe6adu

Accelerated Machine Learning as a Service for Particle Physics Computing

Javier Duarte, Burt Holzman, Sergo Jindariani, Thomas Klijnsma, Benjamin Kreis, Mia Liu, Kevin Pedro, Nhan Tran, Aristeidis Tsaris, Phil Harris, Dylan Rankin, Vladimir Loncar (+12 others)
2019 Zenodo  
Using Microsoft Azure Machine Learning deploying Intel FPGAs to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) milliseconds with our experimental physics  ...  New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains.  ...  Across big science, such as cosmology and large astrophysical surveys, similar trends exist as the experiments grow and the data rates increase.  ... 
doi:10.5281/zenodo.3895029 fatcat:m43b2enphjfitjvofwgyp5mcw4

CEAZ: Accelerating Parallel I/O Via Hardware-Algorithm Co-Designed Adaptive Lossy Compression [article]

Chengming Zhang, Sian Jin, Tong Geng, Jiannan Tian, Ang Li, Dingwen Tao
2021 arXiv   pre-print
Finally, we evaluate CEAZ on five real-world datasets with both a single FPGA board and 128 nodes (to accelerate parallel I/O).  ...  In this paper, we propose a hardware-algorithm co-design for an efficient and adaptive lossy compressor for scientific data on FPGAs (called CEAZ), which is the first lossy compressor that can achieve  ...  Parallel I/O Accelerator Many scientific applications, such as cosmology simulations, need to periodically dump a huge amount of raw simulation data to the storage for post-hoc analysis and visualization  ... 
arXiv:2106.13306v2 fatcat:42fvquu3trcgxncxwdl5izksra

Introduction to the Special Issue on Digital Signal Processing in Radio Astronomy

D. C. Price, J. Kocz, M. Bailes, L. J. Greenhill
2016 Journal of Astronomical Instrumentation  
In particular, graphics processing units (GPUs) and field programmable gate arrays (FPGAs) are being used in place of application-specific circuits (ASICs); high-speed Ethernet and Infiniband are being  ...  Further, to lower hurdles in digital engineering, communities have designed and released general-purpose FPGA-based DSP systems, such as the CASPER ROACH board, ASTRON Uniboard and CSIRO Redback board.  ...  FPGAs excel at low-level DSP tasks with high data throughput requirements and are easily interfaced with ADCs, DACs, and interconnects.  ... 
doi:10.1142/s2251171716020025 fatcat:okeecsvzzvhbxjomvyyojro5yq

Applications and Techniques for Fast Machine Learning in Science [article]

Allison McCarn Deiana, Joshua Agar, Michaela Blott, Giuseppe Di Guglielmo, Javier Duarte, Philip Harris, Scott Hauck, Mia Liu, Mark S. Neubauer, Jennifer Ngadiuba, Seda Ogrenci-Memik, Maurizio Pierini (+74 others)
2021 arXiv   pre-print
loop to accelerate scientific discovery.  ...  In this community review report, we discuss applications and techniques for fast machine learning (ML) in science -- the concept of integrating power ML methods into the real-time experimental data processing  ...  This is true both in online data taking and offline data analysis.  ... 
arXiv:2110.13041v1 fatcat:cvbo2hmfgfcuxi7abezypw2qrm

Axel

Kuen Hung Tsoi, Wayne Luk
2010 Proceedings of the 18th annual ACM/SIGDA international symposium on Field programmable gate arrays - FPGA '10  
Axel contains a collection of nodes; each node can include multiple types of accelerators such as FPGAs (Field Programmable Gate Arrays) and GPUs (Graphics Processing Units).  ...  The Axel system enables the first demonstration of FPGAs, GPUs and CPUs running collaboratively for N-body simulation.  ...  Various applications including molecular dynamics (NAMD), weather modeling (WRF), cosmology data analysis (TPACF) have been implemented on the QP system.  ... 
doi:10.1145/1723112.1723134 dblp:conf/fpga/TsoiL10 fatcat:qyvn4ea3avdqxdncjrdeosoe5m

Mining Discriminative K-mers in DNA Sequences Using Sketches and Hardware Acceleration

Antonio Saavedra, Hans Lehnert, Cecilia Hernandez, Gonzalo Carvajal, Miguel Figueroa
2020 IEEE Access  
Extracting discriminative k-mers is an important and challenging problem in DNA sequence analysis with applications in metagenomics and motif discovery.  ...  More importantly, we designed a custom FPGA-based accelerator for our algorithm on a Xilinx KCU1500 board, which achieves speedups above 78x with the largest datasets, compared to our parallel software  ...  Beyond FPGA-based systems, authors in [35] use GPUs to accelerate halo-finding in cosmological N-body simulation data using sketches.  ... 
doi:10.1109/access.2020.3003918 fatcat:zuxcxgfamzfvrdqnjsyntyzggq

FPGA-accelerated machine learning inference as a service for particle physics computing [article]

Javier Duarte, Philip Harris, Scott Hauck, Burt Holzman, Shih-Chieh Hsu, Sergo Jindariani, Suffian Kha, Benjamin Krei, Brian Le, Mia Liu, Vladimir Lončar, Jennifer Ngadiuba, Kevin Pedro (+10 others)
2019 arXiv   pre-print
New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains.  ...  Using Project Brainwave by Microsoft to accelerate the ResNet50 image classification model, we achieve average inference times of 60 (10) milliseconds with our experimental physics software framework using  ...  The authors thank the NOvA collaboration for the use of its Monte Carlo software tools and data and for the review of this manuscript.  ... 
arXiv:1904.08986v1 fatcat:7rxj5a56w5gd3d7gusykxdztse

Gordon Bell finalists II---$158/GFLOPS astrophysical N-body simulation with reconfigurable add-in card and hierarchical tree algorithm

Atsushi Kawai, Toshiyuki Fukushige
2006 Proceedings of the 2006 ACM/IEEE conference on Supercomputing - SC '06  
On our system, we performed a cosmological N-body simulation with 2.1 million particles, which sustained a performance of 15.39 Gflops averaged over 4.33 hours.  ...  The reconfigurable add-in card houses one FPGA chip, into which we integrated 16 pipeline processors specialized for gravitational force calculation.  ...  KFCR GRAPE-7 is available at the price of 186k JYE, which includes configuration data for logic design of FPGA chip. The price of the host PC is 94k JYE. The overall cost is 2,384 USD.  ... 
doi:10.1145/1188455.1188505 dblp:conf/sc/KawaiF06 fatcat:n4oue7mi2zc5lkr4cylsvhyb4q

FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing

Javier Duarte, Philip Harris, Scott Hauck, Burt Holzman, Shih-Chieh Hsu, Sergo Jindariani, Suffian Khan, Benjamin Kreis, Brian Lee, Mia Liu, Vladimir Lončar, Jennifer Ngadiuba (+11 others)
2019 Computing and Software for Big Science  
New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains.  ...  Using Project Brainwave by Microsoft to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) ms with our experimental physics software framework using Brainwave  ...  The authors thank the NOvA collaboration for the use of its Monte Carlo software tools and data and for the review of this manuscript.  ... 
doi:10.1007/s41781-019-0027-2 fatcat:gutivlorajcdfdlqaabibsoiaa

Smooth Particle Hydrodynamics: Models, Applications, and Enabling Technologies [article]

Piet Hut, Lars Hernquist, George Lake, Jun Makino, Steve McMillan, and Thomas Sterling
1997 arXiv   pre-print
of reconfigurable structures, based on Field Programmable Gate Arrays (FPGAs).  ...  at each stage in this decomposition, by using enabling hardware technology to accelerate the performance of general purpose computers.  ...  • Data Flow -What are the data flow rates across the SPH interface and how does this scale with problem size and with respect to other computational components (N -body simulation, etc.) • Implementation  ... 
arXiv:astro-ph/9710212v1 fatcat:m4hprtirtbfa3al4wszzxd7gci

Real-time data analysis model at LHC and connections to other experiments and fields

Arantza Oyanguren, C. Doglioni, D. Kim, G.A. Stewart, L. Silvestris, P. Jackson, W. Kamleh
2020 EPJ Web of Conferences  
With the upcoming increase of proton-proton collision rates at the Large Hadron Collider (LHC) experiments, and the corresponding increase of data volumes, real-time analysis becomes a key ingredient to  ...  Similar challenges have to be faced in other fields, such as astronomy and cosmology, and I will comment about them.  ...  Using accelerators Hardware accelerator platforms such as GPUs (Graphic Processing Units) or FPGAs are promising architectures for speeding up object reconstruction and will be implemented in the trigger  ... 
doi:10.1051/epjconf/202024511005 fatcat:j6yhobeiancx5d2xacisdbvinm

FPGA-based Acceleration of FT Convolution for Pulsar Search Using OpenCL [article]

Haomiao Wang, Prabu Thiagaraj, Oliver Sinnen
2018 arXiv   pre-print
This paper focuses on the FPGA-based acceleration of the Frequency-Domain Acceleration Search module, which is a part of SKA pulsar search engine.  ...  The performance and power consumption are evaluated using multiple FPGA devices simultaneously and compared with GPU results, which is achieved by porting FPGA-based OpenCL kernels.  ...  ACKNOWLEDGMENTS The authors acknowledge discussions with the TDT, a collaboration between Manchester and Oxford Universities, and MPIfR Bonn and the work benefitted from their collaboration.  ... 
arXiv:1805.12280v2 fatcat:4wlpaig3hndlrmb4prz4pquckq

Accelerating cosmological data analysis with graphics processors

Dylan W. Roeh, Volodymyr V. Kindratenko, Robert J. Brunner
2009 Proceedings of 2nd Workshop on General Purpose Processing on Graphics Processing Units - GPGPU-2  
FPGAs Measured features/ parameters SRC-6 host 2.8 GHz Xeon SRC-6 dual- MAP SGI Altix host 1.4 GHz Itanium 2 RC100 blade # CPUs 2 2 # FPGAs 4 2 # of compute engines 1 17 2  ...  ); computeDD(data, npd, data, npd, 1,binb, nbins, njks, DD); for (i = 0; i < random_count; i++) // loop through random data files { loadRandomData(random[i]); computeRR(random[i], npr[i], random[  ... 
doi:10.1145/1513895.1513896 dblp:conf/asplos/RoehKB09 fatcat:krzebqpkbjhxtn3trqhqjpw2zy
« Previous Showing results 1 — 15 out of 613 results