Filters








15,399 Hits in 9.6 sec

Dynamic parallelization of recursive code

Charlotte Herzeel, Pascal Costanza
2010 Proceedings of the ACM international conference on Object oriented programming systems languages and applications - OOPSLA '10  
We introduce the continuator concept, a runtime structure that tracks and manages the control dependences between such concurrently spawned tasks, ensuring adherence to the sequential semantics of the  ...  While most approaches to automatic parallelization focus on compilation approaches for parallelizing loop iterations, we advocate the need for new virtual machines that can parallelize the execution of  ...  It is truly an amazing piece of technology that sets a very high standard for multiprocessing support in Lisp and other dynamic languages.  ... 
doi:10.1145/1869459.1869491 dblp:conf/oopsla/HerzeelC10 fatcat:idxipwjpcfdddfh6dvn3bilcxe

Compiler generation for interactive graphics using intermediate code [chapter]

Scott Draves
1996 Lecture Notes in Computer Science  
This paper describes a compiler generator (cogen) designed for interactive graphics, and presents preliminary results of its application to pixel-level code.  ...  The cogen accepts and produces a reflective intermediate code in continuationpassing, closure-passing style.  ...  I would like to thank Olivier Danvy, Nick Thompson, and the anonymous reviewers for their comments on drafts of this paper, and Peter Lee for his continuing feedback, faith, and support.  ... 
doi:10.1007/3-540-61580-6_6 fatcat:5hh6fkbrlfgnxfekxk2lj46cs4

URSI National Committee Report: Commission 6. Radio Waves and Circuits. Part 1. Information Theory and Coding

P. Elias
1960 Journal of Research of the National Bureau of Standards Section D Radio Propagation  
The introduction of two thresholds rather than one in a continuous channel introduces a null zone.  ...  Similar results have I been obtained by Shannon for more~'general channels. 671 This work is not yet published, but the case of a continuous channel with additive Gaussian noise has been treated in detail  ...  All alLernative ftpproach to problems of scattering and propagation in random media is Lo work with a perturbc·d con till u um -see papers of Ei nsLein [1910], SlHolu ehowski [1908], Pekeris [1947] ,  ... 
doi:10.6028/jres.064d.076 fatcat:as4o4awtpzaevjutixsygjqbpm

Data-Driven Concurrency for High Performance Computing

George Matheou, Paraskevas Evripidou
2017 ACM Transactions on Architecture and Code Optimization (TACO)  
In this work, we utilize dynamic dataflow/data-driven techniques to improve the performance of high performance computing (HPC) systems.  ...  The proposed framework is based on data-driven multithreading (DDM), a hybrid control-flow/dataflow model that schedules threads based on data availability on sequential processors.  ...  ACKNOWLEDGMENTS The authors would like to thank the Cyprus Institute and the CyTera HPC facility for providing the hardware resources (Intel nodes) used in the evaluation of this work.  ... 
doi:10.1145/3162014 fatcat:nnrbtfrog5hp5ptlzglxuc75gq

RE-Analyzer: From source code to structured analysis

A. B. O'Hare, E. W. Troan
1994 IBM Systems Journal  
The single control process manages 118 O'HARE AND TROAN IBM SYSTEMS JOURNAL, VOL 33, NO 1, 1994 Figure 5 The decomposition of process "GenerateReport.2" Data Flow Diagram: GenerateReport.2 0 1 FILEPtr  ...  for the code analysis part of the RE- Analyzer, which has over 17 000 lines of code.  ... 
doi:10.1147/sj.331.0110 fatcat:2cohz5goeve4jhpzufnx3wlhie

Dynamic compilation of data-parallel kernels for vector processors

Andrew Kerr, Gregory Diamos, S. Yalamanchili
2012 Proceedings of the Tenth International Symposium on Code Generation and Optimization - CHO '12  
while tolerating control-flow divergence.  ...  In particular, exploiting software parallelism without sacrificing portability across the spectrum of commodity and multi-core SIMD processors remains elusive.  ...  ACKNOWLEDGEMENTS The authors would like to thank Vinod Grover, Nathan Clark, and Scott Mahlke for discussions relating to dynamic compilation for vector architectures and to the anonymous reviewers for  ... 
doi:10.1145/2259016.2259020 dblp:conf/cgo/KerrDY12 fatcat:7gpks5jk5zhdjkfqhtvcvq34we

Towards Automatic High-Level Code Deployment on Reconfigurable Platforms: A Survey of High-Level Synthesis Tools and Toolchains

Mostafa W. Numan, Braden J. Phillips, Gavin S. Puddy, Katrina Falkner
2020 IEEE Access  
Dividing an application into kernels and a manager enables logic to be deeply pipelined without control flow hazards. FIGURE 17 illustrates the flow and components of the MaxCompiler.  ...  The introduction of HLS in the design flow reduces the transition from the HLL code to the start of the automatic process, as can be seen from the design flow in FIGURE 1.  ... 
doi:10.1109/access.2020.3024098 fatcat:hk7s2deq6zgp5fnuwvm5k6jodu

Helium: lifting high-performance stencil kernels from stripped x86 binaries to halide DSL code

Charith Mendis, Jeffrey Bosboom, Kevin Wu, Shoaib Kamil, Jonathan Ragan-Kelley, Sylvain Paris, Qin Zhao, Saman Amarasinghe
2015 SIGPLAN notices  
The original optimized code for kernels in stripped binaries is nearly impossible to analyze statically. Instead, we rely on dynamic traces to regenerate the kernels.  ...  We manually rejuvenated Photoshop by replacing eleven of Photoshop's filters with our lifted implementations, giving 1.12× speedup without affecting the user experience.  ...  , and Jeff Chien of the Photoshop team for their helpful input.  ... 
doi:10.1145/2813885.2737974 fatcat:apti6ahwtvh4laxjjvschayrl4

Verification of safety requirements for program code using data abstraction

Frank Stappers, Michel Reniers
2009 Electronic Communications of the EASST  
This paper describes and illustrates a method by which systems with a high degree of parallelism can be verified.  ...  The derived model, consisting of interface calls between processes, is checked for various safety properties with the mCRL2 tool set.  ...  Acknowledgements This work is supported as part of the ITEA project Twins 05004.  ... 
doi:10.14279/tuj.eceasst.23.311 dblp:journals/eceasst/StappersR09 fatcat:aunokcv6i5g7hmrrbp3zoa2vh4

Code and the Transduction of Space

Rob Kitchin, Martin Dodge
2005 Annals of the Association of American Geographers  
The nature of space transduced by code is detailed and illustrated with respect to domestic living, work, communication, transport, and consumption.  ...  The effects of software (code) on the spatial formation of everyday life are best understood through a theoretical framework that utilizes the concepts of technicity (the productive power of technology  ...  We also acknowledge the helpful comments of Audrey Kobayashi and the Annals referees. Notes  ... 
doi:10.1111/j.1467-8306.2005.00454.x fatcat:yy6zzxz24fej5j23mfbcie7vxy

Helium: lifting high-performance stencil kernels from stripped x86 binaries to halide DSL code

Charith Mendis, Jeffrey Bosboom, Kevin Wu, Shoaib Kamil, Jonathan Ragan-Kelley, Sylvain Paris, Qin Zhao, Saman Amarasinghe
2015 Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation - PLDI 2015  
The original optimized code for kernels in stripped binaries is nearly impossible to analyze statically. Instead, we rely on dynamic traces to regenerate the kernels.  ...  We manually rejuvenated Photoshop by replacing eleven of Photoshop's filters with our lifted implementations, giving 1.12× speedup without affecting the user experience.  ...  , and Jeff Chien of the Photoshop team for their helpful input.  ... 
doi:10.1145/2737924.2737974 dblp:conf/pldi/MendisBWKRPZA15 fatcat:omz4dgllc5erbicbio7kqi7f2q

The Kremlin Oracle for Sequential Code Parallelization

Saturnino Garcia, Donghwan Jeon, Christopher Louie, Michael Bedford Taylor
2012 IEEE Micro  
Acknowledgments This research was funded in part by the US National Science Foundation (NSF) under CAREER Award 0846152; by NSF Awards 0725357, 0846152, and 1018850; and by a gift from Advanced Micro Devices  ...  He completed the work described in Figure 1 . 1 Interacting with Kremlin. Figure 2 . 2 Kremlin's structure.  ...  A sample Kremlin session Kremlin uses a simple usage model inspired by the GNU profiler (gprof). Figure 1 demonstrates the user's interaction with Kremlin.  ... 
doi:10.1109/mm.2012.52 fatcat:aqcvratjgvbolols5q2ct6kqp4

Tiled-MapReduce

Rong Chen, Haibo Chen
2013 ACM Transactions on Architecture and Code Optimization (TACO)  
The prevalence of chip multiprocessors opens opportunities of running data-parallel applications originally in clusters on a single machine with many cores.  ...  TMR partitions a large MapReduce job into a number of small subjobs and iteratively processes one subjob at a time with efficient use of resources; TMR finally merges the results of all subjobs for output  ...  Execution Flow The top part of Figure 2 illustrates the overall execution flow of a Tiled-MapReduce job and the implementation of Tiled-MapReduce runtime.  ... 
doi:10.1145/2445572.2445575 fatcat:fbfbnro6rzegfb4y5vro3i4zva

Bootstrapping MDE development from ROS manual code: Part 2—Model generation and leveraging models at runtime

Nadia Hammoudeh García, Harshavardhan Deshpande, André Santos, Björn Kahl, Mirko Bordignon
2021 Journal of Software and Systems Modeling  
In this paper, we present two methods based on metamodels for automated generation of models from manually written artifacts: (1) through static code analysis and (2) by monitoring the execution of a running  ...  For both methods, we present tools that leverage the potentials of our contributions, with a special focus on their application at runtime to observe and diagnose a real system during its execution.  ...  run by the German Federal Ministry for Economic Affairs and Energy and is managed by the DLR project management agency "Gesellschaft, Innovation, Technologie-Informationstechnologien/Elektromobilität"  ... 
doi:10.1007/s10270-021-00873-2 fatcat:624ilug56jdhjebfktbhaw2wjq

The Arepo public code release [article]

Rainer Weinberger, Volker Springel, Rüdiger Pakmor
2020 arXiv   pre-print
The employed parallelization algorithms of Arepo are deterministic and produce binary-identical results when re-run on the same machine and with the same number of MPI ranks.  ...  This allows simulations with high dynamic range to be performed efficiently.  ...  The authors would like to thank the full user base of Arepo for their continued encouragement to realize a public release of the code, and for their long-standing efforts in putting the code to great scientific  ... 
arXiv:1909.04667v2 fatcat:dt5ujxdlejhlvizzdmpfq6afla
« Previous Showing results 1 — 15 out of 15,399 results