Filters








9,665 Hits in 8.0 sec

The potential for using thread-level data speculation to facilitate automatic parallelization

J.G. Steffan, T.C. Mowry
Proceedings 1998 Fourth International Symposium on High-Performance Computer Architecture  
This paper explores the potential for using thread-level data speculation (TLDS) to overcome this limitation by allowing the compiler to view parallelization solely as a cost/benefit tradeoff, rather than  ...  the primary data caches to buffer speculative state.  ...  Conclusions To enable a potential breakthrough in the compiler's ability to automatically parallelize non-numeric applications, we have investigated thread-level data speculation (TLDS)-a technique which  ... 
doi:10.1109/hpca.1998.650541 dblp:conf/hpca/SteffanM98 fatcat:in5m6iv35fbbfnvevvjzfxyoly

The Jrpm system for dynamically parallelizing Java programs

Michael K. Chen, Kunle Olukotun
2003 Proceedings of the 30th annual international symposium on Computer architecture - ISCA '03  
CMPs have low sharing and communication costs relative to traditional multiprocessors, and thread-level speculation (TLS) simplifies program parallelization by allowing us to parallelize optimistically  ...  We describe the Java runtime parallelizing machine (Jrpm), a complete system for parallelizing sequential programs automatically.  ...  These basic components are used to create a system that can automatically compile programs to exploit thread-level parallelism.  ... 
doi:10.1145/859618.859668 fatcat:quf7z5ocfffjdnkxdhjyy7hvya

The Jrpm system for dynamically parallelizing Java programs

Michael K. Chen, Kunle Olukotun
2003 SIGARCH Computer Architecture News  
CMPs have low sharing and communication costs relative to traditional multiprocessors, and thread-level speculation (TLS) simplifies program parallelization by allowing us to parallelize optimistically  ...  We describe the Java runtime parallelizing machine (Jrpm), a complete system for parallelizing sequential programs automatically.  ...  These basic components are used to create a system that can automatically compile programs to exploit thread-level parallelism.  ... 
doi:10.1145/871656.859668 fatcat:7paj5trthrfifaxkw3kuk2wquy

The Jrpm system for dynamically parallelizing Java programs

Michael K. Chen, Kunle Olukotun
2003 Proceedings of the 30th annual international symposium on Computer architecture - ISCA '03  
CMPs have low sharing and communication costs relative to traditional multiprocessors, and thread-level speculation (TLS) simplifies program parallelization by allowing us to parallelize optimistically  ...  We describe the Java runtime parallelizing machine (Jrpm), a complete system for parallelizing sequential programs automatically.  ...  These basic components are used to create a system that can automatically compile programs to exploit thread-level parallelism.  ... 
doi:10.1145/859666.859668 fatcat:imo5rgynjfektfdplhdkh5l4oa

Speculative Multithreaded Processors [chapter]

Gurindar S. Sohi, Amir Roth
2000 Lecture Notes in Computer Science  
We expect to see two major forms of thread-level speculation: control-driven and data-driven.  ...  Multithreaded architectures provide new opportunities for extracting parallelism from a single program via thread level speculation.  ...  Acknowledgements This work was supported in part by National Science Foundation grants MIP-9505853 and CCR-9900584, donations from Intel and Sun Microsystems, the University of Wisconsin Graduate School  ... 
doi:10.1007/3-540-44467-x_23 fatcat:2ecvlteawzb57ek3hz57axiejy

Speculative multithreaded processors

G.S. Sohi, A. Roth
2001 Computer  
Such architectures can extract parallelism from a sequential program via thread-level speculation-be it control-driven or data-driven-giving them the flexibility to operate in both multiple-program, high-throughput  ...  The motivation for using speculative multithreading comes from two directions: On the one hand, we are already witnessing the diminishing potential of current techniques to extract parallelism from single  ...  Acknowledgments This work was supported in part by National Science Foundation grants MIP-9505853, CCR-9900584 and 0071924, donations from Intel and Sun Microsystems, the University of Wisconsin Graduate  ... 
doi:10.1109/2.917542 fatcat:fsdtjvtfsnheljrwckngvqg73m

Compiler-Driven Dependence Profiling to Guide Program Parallelization [chapter]

Peng Wu, Arun Kejariwal, Călin Caşcaval
2008 Lecture Notes in Computer Science  
Thread-level speculation (TLS) has been proposed as a technique to parallelize the execution of serial codes or serial sections of parallel codes.  ...  We used DProf to quantify the potential of this approach and we present results on selected applications from the SPEC CPU2006 and SEQUOIA benchmarks.  ...  Introduction Thread-level speculation (TLS) [28, 16, 26, 12] is one technique that has been proposed for parallelizing sequential codes to exploit parallel and multi-core architectures.  ... 
doi:10.1007/978-3-540-89740-8_16 fatcat:u3mqosq6cneqvlx4mp5r54xgou

Exposing speculative thread parallelism in SPEC2000

Manohar K. Prabhu, Kunle Olukotun
2005 Proceedings of the tenth ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '05  
One way to enhance the performance of chip multiprocessors that has received considerable attention is the use of thread-level speculation (TLS).  ...  For each application, we discuss how and where parallelism was located within the application, the impediments to extracting this parallelism using TLS, and the code transformations that were required  ...  The authors would also like to thank Lance Hammond for extensive discussions and key insights on this paper and for support of the Hydra CMP simulator.  ... 
doi:10.1145/1065944.1065964 dblp:conf/ppopp/PrabhuO05 fatcat:jrl5krbtmfb7jdga4sgfuxqhxi

A Survey on Thread-Level Speculation Techniques

Alvaro Estebanez, Diego R. Llanos, Arturo Gonzalez-Escribano
2016 ACM Computing Surveys  
Thread-Level Speculation (TLS) is a promising technique that allows the parallel execution of sequential code without relying on a prior, compile-time dependence analysis.  ...  In this work we introduce the technique, present a taxonomy of TLS solutions, and summarize and put into perspective the most relevant advances in this field.  ...  COST Program Action IC1305: Network for Sustainable Ultrascale Computing (NESUS).  ... 
doi:10.1145/2938369 fatcat:yqqyjoaidvci3d4dyuw2jc2p2i

Speculative Decoupled Software Pipelining

Neil Vachharajani, Ram Rangan, Easwaran Raman, Matthew J. Bridges, Guilherme Ottoni, David I. August
2007 Parallel Architecture and Compilation Techniques (PACT), Proceedings of the International Conference on  
By speculating past infrequent dependences, the benefit of DSWP is increased by making it applicable to more loops, facilitating better balanced threads, and enabling parallelized loops to be run on more  ...  To avoid burdening programmers with the responsibility of parallelizing their applications, some researchers have advocated automatic thread extraction.  ...  Acknowledgments We thank the entire Liberty Research Group for their support and feedback during this work. Additionally, we thank the anonymous reviewers for their insightful comments.  ... 
doi:10.1109/pact.2007.4336199 fatcat:zdfi2dh3ujcsvlshzwhwcnejbi

Runtime automatic speculative parallelization

Ben Hertzberg, Kunle Olukotun
2011 International Symposium on Code Generation and Optimization (CGO 2011)  
In contrast to other systems for automatic speculative parallelization, RASP uses dynamic binary translation to optimize applications on-the-fly, without any need for recompilation or source code.  ...  We present Runtime Automatic Speculative Parallelization (RASP), a technique for the dynamic extraction of speculative threads from a running application in a user-transparent fashion.  ...  SUMMARY We evaluate the potential for runtime automatic speculative parallelization of sequential code.  ... 
doi:10.1109/cgo.2011.5764675 dblp:conf/cgo/HertzbergO11 fatcat:z4qldi2pbnfrfhojrcwoa2xvom

Function level parallelism driven by data dependencies

Sean Rul, Hans Vandierendonck, Koen De Bosschere
2007 SIGARCH Computer Architecture News  
Applying this framework to sequential programs can teach us how much parallelism is present in a program, but also tells us what the most appropriate parallel construct for a program is, e.g. a pipeline  ...  In this paper, we propose a framework for extracting potential parallelism from programs.  ...  Acknowledgments We would like to thank the anonymous reviewers for their valuable feedback.  ... 
doi:10.1145/1241601.1241612 fatcat:fi7lwokainapdmz4bl2l6cetnm

Loop Selection for Thread-Level Speculation [chapter]

Shengyue Wang, Xiaoru Dai, Kiran S. Yellajyosula, Antonia Zhai, Pen-Chung Yew
2006 Lecture Notes in Computer Science  
Thread-level speculation (TLS) allows potentially dependent threads to speculatively execute in parallel, thus making it easier for the compiler to extract parallel threads.  ...  for extracting parallel threads.  ...  Thread-level speculation (TLS) [3, 6, 9, 11, 16, 22, 24, 26] facilitates the parallelization of such applications by allowing potentially dependent threads to execute in parallel while maintaining the  ... 
doi:10.1007/978-3-540-69330-7_20 fatcat:pudo5nhm4rawligrqbksty2acu

Tight analysis of the performance potential of thread speculation using spec CPU 2006

Arun Kejariwal, Xinmin Tian, Milind Girkar, Wei Li, Sergey Kozhukhov, Utpal Banerjee, Alexander Nicolau, Alexander V. Veidenbaum, Constantine D. Polychronopoulos
2007 Proceedings of the 12th ACM SIGPLAN symposium on Principles and practice of parallel programming - PPoPP '07  
Recently, thread-level speculation (TLS) has been proposed as a means to parallelize difficult-to-analyze serial codes.  ...  In this paper, we present a tight analysis of the (unique) performance potential of both: (a) TLS in general and (b) specific types of thread-level speculation, viz., control speculation, data dependence  ...  Subject to overhead incurred due to spawning of multiple speculative threads and the performance penalty incurred due to misspeculation, procedure-level TLS can potentially yield higher levels of parallelism  ... 
doi:10.1145/1229428.1229475 dblp:conf/ppopp/KejariwalTGLKBNVP07 fatcat:s2oeavxngzes5ajf3or26fydf4

Runtime Parallelization of Static and Dynamic Irregular Array of Array References

Parwat Singh Anjanaa, N. Naga Maruthia, Sagar Gujjunooria, Madhu Orugantib
2018 International Journal of Engineering & Technology  
To detect the dependencies between footprints and for synchronization of threads at runtime, we use bit level operations.  ...  For pro ling the runtime information, shared and private data structures are used.  ...  Since we are using lazy version management for data and conflict detection thread-local data structure will contain the most updated valuewhich will be passed to a more speculative thread using value-speculation  ... 
doi:10.14419/ijet.v7i4.6.20452 fatcat:jprxbbpnhjfntmxn7kz4ymf3gq
« Previous Showing results 1 — 15 out of 9,665 results