Filters








42,122 Hits in 4.2 sec

High-performance parallel graph reduction [chapter]

Simon L Peyton Jones, Chris Clack, Jon Salkild
1989 Lecture Notes in Computer Science  
This paper outlines some of the issues raised by parallel compiled graph reduction, and presents the approach we have adopted for our parallel machine, GRIP.  ...  Parallel graph reduction is an attractive implementation for functional programming languages because of its simplicity and inherently distributed nature.  ...  Compiled graph reduction Our first model for parallel reduction was an interpretive one called the four-stroke reduction engineCtac86a, and we now have a running parallel implementation of this machine  ... 
doi:10.1007/3540512845_40 fatcat:rhom626ekvap7kq3dnjqmapfsi

Implementation of speculative parallelism in functional languages

P.V.R. Murthy, V. Rajaraman
1994 IEEE Transactions on Parallel and Distributed Systems  
This provides a basis for identifying useful speculative parallelism in a program.  ...  The performance of speculative evaluation is compared with that of lazy evaluation, and the necessary conditions under which speculative evaluation performs better are identified.  ...  Sastry for their many helpful discussions.  ... 
doi:10.1109/71.329669 fatcat:r4gv6u6oifezbddnnwpadgnlqu

Implementation of Sensitivity Analysis for Automatic Parallelization [chapter]

Silvius Rus, Maikel Pennings, Lawrence Rauchwerger
2008 Lecture Notes in Computer Science  
., privatization or pushback parallelization, and static and dynamic evaluation of complex conditions for loop parallelization.  ...  We concern ourselves with multi-version and parallel code generation as well as the use of speculative parallelization when other, less costly options fail.  ...  These predicates are returned to the programmer for evaluation (for interactive compilation).  ... 
doi:10.1007/978-3-540-89740-8_22 fatcat:66kx5qzjcbbexkeyd4ztkgqml4

The four-stroke reduction engine

Chris Clack, Simon L. Peyton Jones
1986 Proceedings of the 1986 ACM conference on LISP and functional programming - LFP '86  
This paper presents an algorithm for the parallel graph reduction of a functional program.  ...  Functional languages offer a powerful lever on the programming of parallel machines, and the most promising model for implementing these languages is graph reduction.  ...  A model for parallel graph reduction Our proposal for parallel graph reduction has the following features: ( i ) The reduction of the graph is performed by the concurrent execution of many tasks, each  ... 
doi:10.1145/319838.319865 fatcat:75lx6wrkjbaylbyzqau7v7mfsu

Exploitation of nested thread-level speculative parallelism on multi-core systems

Arun Kejariwal, Milind Girkar, Xinmin Tian, Hideki Saito, Alexandru Nicolau, Alexander V. Veidenbaum, Utpal Banerjee, Constantine D. Polychronopoulos
2010 Proceedings of the 7th ACM international conference on Computing frontiers - CF '10  
Nested thread-level speculative parallelization has been proposed as a means to exploit the hardware parallelism of such systems.  ...  In this paper, we present a methodology to gauge the efficacy of nested thread-level speculation with increasing level of nesting.  ...  BACKGROUND In this section, we present an overview of the speculative execution model and the basics of conditional probability. 1 The need for exploiting nested speculative thread-level parallelism  ... 
doi:10.1145/1787275.1787302 dblp:conf/cf/KejariwalGTSNVBP10 fatcat:urckqjih7veqnjqn7cvvydsuvq

Why Parallel Functional Programming Matters: Panel Statement [chapter]

Kevin Hammond
2011 Lecture Notes in Computer Science  
The evaluate-and-die technique improves on this by taking advantage of graph reduction. If a sparked node is needed (not speculative), then it must be attached to the parent computation.  ...  The novel graph reducer GRIP (Graph Reduction in Parallel) 102, 101] is built from a network of distributed conventional processors.  ... 
doi:10.1007/978-3-642-21338-0_17 fatcat:n3tvygfvdbcrpmmbrnxn7xshcq

Discovery and exploitation of general reductions: A constraint based approach

Philip Ginsbach, Michael F. P. O'Boyle
2017 2017 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)  
Once discovered, we automatically generate parallel code to exploit the reduction. This approach is robust and was evaluated on C versions of well known benchmark suites: NAS, Parboil and Rodinia.  ...  Discovering and exploiting scalar reductions in programs has been studied for many years. The discovery of more complex reduction operations has, however, received less attention.  ...  Candidates for speculative parallelization are determined by searching for update-chains in the data flow graph.  ... 
doi:10.1109/cgo.2017.7863746 fatcat:6f2ahogcc5bgtili2xiggy2w3a

Exploiting coarse-grain speculative parallelism

Hari K. Pyla, Calvin Ribbens, Srinidhi Varadarajan
2011 SIGPLAN notices  
Speculative execution at coarse granularities (e.g., codeblocks, methods, algorithms) offers a promising programming model for exploiting parallelism on modern architectures.  ...  Anumita can be used to improve the performance of hard to parallelize algorithms whose performance is highly dependent on input data.  ...  During reduction, conflicting subgraphs are recolored. Despite such efforts, the challenge still persists to develop efficient parallel algorithms for vertex coloring.  ... 
doi:10.1145/2076021.2048110 fatcat:rsyjze662nh4pf62mi26rirdwu

Exploiting coarse-grain speculative parallelism

Hari K. Pyla, Calvin Ribbens, Srinidhi Varadarajan
2011 Proceedings of the 2011 ACM international conference on Object oriented programming systems languages and applications - OOPSLA '11  
Speculative execution at coarse granularities (e.g., codeblocks, methods, algorithms) offers a promising programming model for exploiting parallelism on modern architectures.  ...  Anumita can be used to improve the performance of hard to parallelize algorithms whose performance is highly dependent on input data.  ...  During reduction, conflicting subgraphs are recolored. Despite such efforts, the challenge still persists to develop efficient parallel algorithms for vertex coloring.  ... 
doi:10.1145/2048066.2048110 dblp:conf/oopsla/PylaRV11 fatcat:tgatnmdn3zekhmo4say74qibua

Exclusive squashing for thread-level speculation

Álvaro García-Yágüez, Diego R. Llanos, Arturo González-Escribano
2011 Proceedings of the 20th international symposium on High performance distributed computing - HPDC '11  
Speculative parallelization is a runtime technique that optimistically executes sequential code in parallel, checking that no dependence violations appear.  ...  Results show a reduction of 38.5% to 81.8% in the number of restarted threads for real application loops and up to a 10% speedup, depending on the amount of local computation.  ...  Belén Palop for many helpful discussions on this topic.  ... 
doi:10.1145/1996130.1996172 dblp:conf/hpdc/Garcia-YaguezFG11 fatcat:tq66jmp3afdh3jwjkr2xzwimeq

GUM

P. W. Trinder, K. Hammond, J. S. Mattson, A. S. Partridge, S. L. Peyton Jones
1996 Proceedings of the ACM SIGPLAN 1996 conference on Programming language design and implementation - PLDI '96  
For example GUM is available by FTP for a Sun SPARCserver multiprocessor and for a networks of Sun SPARC workstations.  ...  GUM is a portable, parallel implementation of the Haskell functional language which has been publicly released with version 0.26 of the Glasgow Haskell Compiler (GHC).  ...  We also expect to need to tune our system, especially for shared-memory systems, and perhaps introduce new parallel hints that can be exploited by some classes of architecture.  ... 
doi:10.1145/231379.231392 dblp:conf/pldi/TrinderHMPJ96 fatcat:5gnb2fozw5g6bbdum762372ece

GUM

P. W. Trinder, K. Hammond, J. S. Mattson, A. S. Partridge, S. L. Peyton Jones
1996 SIGPLAN notices  
For example GUM is available by FTP for a Sun SPARCserver multiprocessor and for a networks of Sun SPARC workstations.  ...  GUM is a portable, parallel implementation of the Haskell functional language which has been publicly released with version 0.26 of the Glasgow Haskell Compiler (GHC).  ...  We also expect to need to tune our system, especially for shared-memory systems, and perhaps introduce new parallel hints that can be exploited by some classes of architecture.  ... 
doi:10.1145/249069.231392 fatcat:2mdtccygy5gf3oirv3hhuk6vxa

A provably time-efficient parallel implementation of full speculation

John Greiner, Guy E. Blelloch
1996 Proceedings of the 23rd ACM SIGPLAN-SIGACT symposium on Principles of programming languages - POPL '96  
Speculative evaluation, including leniency and futures, is often used to produce high degrees of parallelism.  ...  For this purpose we consider a simple language based on speculative evaluation.  ...  INTRODUCTION Futures, lenient languages, and several implementations of graph reduction for lazy languages all use speculative evaluation (call-by-speculation [Hudak and Anderson 1987] ) to expose parallelism  ... 
doi:10.1145/237721.237797 dblp:conf/popl/GreinerB96 fatcat:ygckgzim6jh75cfwgw743jnz6i

A provably time-efficient parallel implementation of full speculation

John Greiner, Guy E. Blelloch
1999 ACM Transactions on Programming Languages and Systems  
Speculative evaluation, including leniency and futures, is often used to produce high degrees of parallelism.  ...  For this purpose we consider a simple language based on speculative evaluation.  ...  INTRODUCTION Futures, lenient languages, and several implementations of graph reduction for lazy languages all use speculative evaluation (call-by-speculation [Hudak and Anderson 1987] ) to expose parallelism  ... 
doi:10.1145/316686.316690 fatcat:mtv7u3o7wnbz5mvkg5rm5ip62y

DA Based Systematic Approach Using Speculative Addition for High Speed DSP Applications

G Reddy Hemantha, S Varadarajan, M N. Giriprasad
2018 International Journal of Engineering & Technology  
The proposed speculative adder based on Han-Carlson parallel-prefix topology attains better latency reduction than variable latency Kogge-Stone topology.  ...  In recent years Parallel-prefix topologies has been emerged to offer a high-speed solution for many DSP applications.  ...  This paper, for the first time, we present speculation with variable latency to link equations via parallel-prefix computation.  ... 
doi:10.14419/ijet.v7i2.24.12030 fatcat:ft4s32mfbjc4fl72jheayhlke4
« Previous Showing results 1 — 15 out of 42,122 results