128 Hits in 2.6 sec

Toward multi-target autotuning for accelerators

Nick Chaimov, Boyana Norris, Allen Malony
2014 2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)  
to changes in architectures and programming models, and enable very specialized optimizations that are not possible or likely to be provided by mainstream compilers.  ...  In this paper we describe the new OpenCL code generator and autotuner OrCL and the introduction of detailed performance measurement into the autotuning process.  ...  INTRODUCTION To overcome the challenges posed by the limits of Dennard scaling, hardware has evolved toward increased levels of parallelism and heterogeneity, whereas prevalent programming models, languages  ... 
doi:10.1109/padsw.2014.7097851 dblp:conf/icpads/ChaimovNM14 fatcat:ezx34pl57zarzpzdmkj4hmvc7i

Performance improvement of distributed systems by autotuning of the configuration parameters

Fan Zhang, Junwei Cao, Lianchen Liu, Cheng Wu
2011 Tsinghua Science and Technology  
There are too many uncertainties affecting the performance autotuning.  ...  This paper introduces an Ordinal Optimization (OO) based strategy combined with a back propagation neural network for autotuning of the configuration parameters.  ...  To make the test more applicable to a real scenario, the workload was not added simultaneously.  ... 
doi:10.1016/s1007-0214(11)70063-3 fatcat:27o2oiv6xjbgxnkdrln4dz4eyy

Autotuning in High-Performance Computing Applications

Prasanna Balaprakash, Jack Dongarra, Todd Gamblin, Mary Hall, Jeffrey K. Hollingsworth, Boyana Norris, Richard Vuduc
2018 Proceedings of the IEEE  
If autotuning is to be widely used in the HPC community, researchers must address the Manuscript  ...  Autotuning has the potential to dramatically improve the performance portability of petascale and exascale applications.  ...  The goal of this paper was twofold: to describe the state of the art in autotuning and to present future requirements to move autotuning toward mainstream use in HPC.  ... 
doi:10.1109/jproc.2018.2841200 fatcat:576fhps52banrfowi6ykk7za6a

Model-Free Current Loop Autotuning for Synchronous Reluctance Motor Drives

Dario Pasqualotto, Fabio Tinazzi, Mauro Zigliotto
2020 Automation  
The present paper uses a relay feedback approach to perform autotuning without resorting to any parameter knowledge.  ...  Introduction In the mainstream of energy and environmental global paradigms, the electric drives technology is moving toward more efficient motors and more intelligent control systems, able to minimize  ...  A well known method, i.e., the relay autotuning method firstly proposed in [7] , can be fruitfully adopted toward this aim.  ... 
doi:10.3390/automation1010003 fatcat:qxlrttyo6rc6zh3dcnontnkxj4


Jason Ansel, Maciej Pacula, Yee Lok Wong, Cy Chan, Marek Olszewski, Una-May O'Reilly, Saman Amarasinghe
2012 Proceedings of the 2012 international conference on Compilers, architectures and synthesis for embedded systems - CASES '12  
Failure to re-autotune programs often leads to widespread use of sub-optimal algorithms.  ...  Modern high performance libraries, such as ATLAS and FFTW, and programming languages, such as PetaBricks, have shown that autotuning computer programs can lead to significant speedups.  ...  By eliminating any extra steps, we believe that SiblingRivalry can bring autotuning to the mainstream program optimization.  ... 
doi:10.1145/2380403.2380425 dblp:conf/cases/AnselPWCOOA12 fatcat:doghqgnt25eqngsaeccktifeii

Energy-Efficient Computing for Extreme-Scale Science

David Donofrio, Leonid Oliker, John Shalf, Michael F. Wehner, Chris Rowen, Jens Krueger, Shoaib Kamil, Marghoob Mohiyuddin
2009 Computer  
The challenge of moving high-performance computing architecture toward exaflops has staggering economic and political ramifications.  ...  It makes more sense therefore to leverage the considerable innovation of the low-power architectures developed for embedded computing markets and design a machine capable of the exaflops performance (1  ...  These rates are based on the computational efficiency rates of current mainstream rates.  ... 
doi:10.1109/mc.2009.353 fatcat:qerwoknemnaivcn2j55l2oc7iu

Contextualizing Hip Hop Sonic Cool Pose in Late Twentieth– and Twenty–first–century Rap Music

Regina Bradley
2017 Current Musicology  
Rapper Shawn "Jay-Z" Carter, for example, sought to attack T-Pain's autotune modification of his voice and music, releasing "D.O.A (Death of Autotune)" on his 2009 album The Blueprint 3.  ...  In a looped sample of Janko Nilvic and Dave Sucky's "In the Space, " Jay-Z dismisses autotune as irrelevant and raps: "this anti-autotune, death of the ringtone/this ain't for iTunes/ this ain't for sing  ... 
doi:10.7916/d8g15zdk fatcat:gz2s5o5bhnawlnhik2zpzf6yqq

Making Room for 21st Century Musicianship in Higher Education

Leah Kardos
2018 Action, Criticism and Theory for Music Education  
Back in 2010, Sufjan Stevens painted a picture of a transformed/purified man through the use of the autotune effect in "Impossible Soul" (from 2010's The Age of Adz); in 2016 artists as mainstream and  ...  Corrective digital tools such as autotune and quantization have entered the cultural language as signifiers, and artists are now using them to make deeper statements and not just to correct imperfections  ...  Making room for 21st century musicianship in higher education. Action, Criticism, and Theory for Music Education 17 (1)  ... 
doi:10.22176/act17.1.33 fatcat:lekxdy3qzjgdjkusqg5vhspozm

A Hardware-Software Blueprint for Flexible Deep Learning Specialization [article]

Thierry Moreau, Tianqi Chen, Luis Vega, Jared Roesch, Eddie Yan, Lianmin Zheng, Josh Fromm, Ziheng Jiang, Luis Ceze, Carlos Guestrin, Arvind Krishnamurthy
2019 arXiv   pre-print
We propose a flow that performs design space exploration to generate a customized hardware architecture and software operator library that can be leveraged by mainstream learning frameworks.  ...  Schedule Exploration for Operator Autotuning Schedule autotuning is the process by which an automated search algorithm attempts to optimize a given program or workload towards peak hardware performance  ...  Additionally, advances in systems show trends towards heterogeneous multi-accelerator systems and scale-out acceleration.  ... 
arXiv:1807.04188v3 fatcat:wpafekkrqzffzfe7vulaa6qnva

Antiwindup-aware PI autotuning

Alessandro Vittorio Papadopoulos, Alberto Leva
2012 IFAC Proceedings Volumes  
As a result, autotuned PIDs may somehow exhibit undesired behaviour in the presence of large transients.  ...  In the past, the mainstream approach has been to avoid accounting for the antiwindup mechanism in the tuning method, and devise antiwindup schemes that provide "reasonably uniform results" whatever tuning  ... 
doi:10.3182/20120328-3-it-3014.00094 fatcat:uzbjugijkfez3otfa5nngx4tf4

Stencil-Aware GPU Optimization of Iterative Solvers

Daniel Lowell, Jeswin Godwin, Justin Holewinski, Deepan Karthik, Chekuri Choudary, Azamat Mametjanov, Boyana Norris, Gerald Sabin, P. Sadayappan, Jason Sarich
2013 SIAM Journal on Scientific Computing  
We also describe autotuning of CUDA implementations based on high-level descriptions of the stencil-based matrix and vector operations.  ...  The work described here is the first necessary step toward enabling the semi-automated creation of libraries tuned for specific applications and target architectures.  ...  The significant exception among mainstream compilers is the open-source GNU Compiler Collection (GCC) [14] and specifically the Milepost GCC component [12] , which employs a machine-learningbased approach  ... 
doi:10.1137/120883153 fatcat:ds2t67ie4bcehh7qsbbomwdogu

Bridging the Performance Gap between Manual and Automatic Compilers with Intent-based Compilation

Waseem Ahmed
2014 International Journal of Computer Applications  
The C file generated by the CI can be further optimized using existing autotuners [5, 27, 28] .  ...  Also, autotuners and many code-to-code transformers insert toolspecific compiler directives and profiling statements within code.  ... 
doi:10.5120/15010-3303 fatcat:7rjc3finfzfpjjvrbjrjufeova

User-Directed Loop-Transformations in Clang [article]

Michael Kruse, Hal Finkel
2018 arXiv   pre-print
Simple transformations such as loop unrolling are already implemented in most mainstream compilers. We recently submitted a proposal to add generalized loop transformations to the OpenMP standard.  ...  Such an autotuner would therefore likely suggest transformations that make the program return wrong results or crash.  ...  This unfortunately makes it hard for an autotuner which only measures the timing difference without understanding the code.  ... 
arXiv:1811.00624v1 fatcat:ssvwhrpwzvbblkinqzpdpvwafi

Song Form and the Mainstreaming of Hip-Hop Music

Ben Duinker
2021 Current Musicology  
This paper explores the nature and evolution of song form in hip-hop music and uses them as a musical lens to view the gradual and ongoing mainstreaming of this genre.  ...  Song form in North American hip-hop music has evolved along the genre's journey from its origins as a live musical practice, through its commercial ascent in the 1980s and 1990s, to its dominance of mainstream  ...  Finally, I investigate how a gradual shift in focus toward verse-hook songs can be understood in relation to hip-hop music's ongoing mainstreaming.  ... 
doi:10.52214/cm.v107i.7177 fatcat:5o32e5bryrdnxeaftbbxdlsnri

FCJ-183 iHootenanny: A Folk Archeology of Social Media

Henry Adam Svec
2015 Fibreculture Journal  
I will ultimately consider how the mediation idealised by the Almanacs has trickled down to a narcissistic will-to-be-'in touch' in mainstream digital culture, making the Hootenanny a virtual path untaken  ...  Cellular phones are the instruments making this recombination possible.  ...  politics in an age of 'communicative capitalism', I will ultimately consider how the techno-authentic mediation idealised by the Almanacs has trickled down to a narcissistic will-to-be-'in touch' in mainstream  ... 
doi:10.15307/fcj.25.183.2015 fatcat:grecbow3ivditjfyphwbwvdtgi
« Previous Showing results 1 — 15 out of 128 results