Filters








534 Hits in 3.1 sec

Thread-spawning schemes for speculative multithreading

P. Marcuello, A. Gonzalez
Proceedings Eighth International Symposium on High Performance Computer Architecture  
Our profile-based spawning scheme is evaluated on a Clustered Speculative Multithreaded Processor and results show large performance benefits.  ...  Speculative multithreading has been recently proposed to boost performance by means of exploiting thread-level parallelism in applications difficult to parallelize.  ...  Acknowledgements This work has been supported by grant CICYT TIC 511/98, The research described in this paper has been developed using the resources of the European Center for Parallelism of Barcelona  ... 
doi:10.1109/hpca.2002.995698 dblp:conf/hpca/MarcuelloG02 fatcat:erbnxt24ffhydb3gox5hvl3lcq

Thread partitioning and value prediction for exploiting speculative thread-level parallelism

P. Marcuello, A. Gonzalez, J. Tubella
2004 IEEE transactions on computers  
In this work, several hardware-based schemes for partitioning the program into speculative threads are analyzed and evaluated.  ...  In general, we find that spawning threads associated to loop iterations is the most effective technique.  ...  The authors would like to thank Bodo Parady (Sun Microsystems, Inc.) for his insightful comments on this paper.  ... 
doi:10.1109/tc.2004.1261823 fatcat:a2euq2as3ra7jihbmtoaw47cgy

Combining thread level speculation helper threads and runahead execution

Polychronis Xekalakis, Nikolas Ioannou, Marcelo Cintra
2009 Proceedings of the 23rd international conference on Conference on Supercomputing - ICS '09  
Generating thread-parallel programs is hard and thread-level speculation (TLS) has been suggested as an execution model that can speculatively exploit thread-level parallelism (TLP) even when thread independence  ...  Each one of these multithreaded execution models works best for different applications and application phases.  ...  For loops, spawn points are placed at the beginning of the loop body, so that each iteration of the loop spawns the next iteration as a speculative thread.  ... 
doi:10.1145/1542275.1542333 dblp:conf/ics/XekalakisIC09 fatcat:qkr7dceoxvbqxcf3jiomghbfxa

Speculative multithreaded processors

Pedro Marcuello, Antonio González, Jordi Tubella
1998 Proceedings of the 12th international conference on Supercomputing - ICS '98  
The selection mechanisms for those speculative threads are referred to as thread-spawning policies.  ...  Different approaches for splitting programs into threads are studied. These mechanisms identify in which parts of the program speculative thread-level parallelism can be effectively exploited.  ...  RELATED WORK The most usual spawning scheme used by speculative multithreaded architectures in the literature is based on spawning speculative threads at loop iterations, either at run-time by means of  ... 
doi:10.1145/277830.277850 dblp:conf/ics/MarcuelloGT98 fatcat:ytextja3bncehcdetvjtb5pvda

Clustered speculative multithreaded processors

Pedro Marcuello, Antonio González
1999 Proceedings of the 13th international conference on Supercomputing - ICS '99  
A main feature of the proposed microarchitecture is its capability to spawn speculative threads from a single-thread application at run-time.  ...  Spawning a speculative thread involves predicting its control flow as well as its dependences with other threads and the values that flow through them.  ...  Acknowledgements We would like to thank José González for his insightful comments on this paper.  ... 
doi:10.1145/305138.305214 dblp:conf/ics/MarcuelloG99 fatcat:4osvinevx5fongtizk2uuvcjje

Mixed speculative multithreaded execution models

Polychronis Xekalakis, Nikolas Ioannou, Marcelo Cintra
2012 ACM Transactions on Architecture and Code Optimization (TACO)  
Mixed speculative multithreaded execution models. ACM Trans.  ...  This technique allows programmers/compilers to generate threads without checking for inter-thread data and control dependences, which are then transparently enforced by the hardware.  ...  For loops, spawn points are placed at the beginning of the loop body, so that each iteration of the loop spawns the next iteration as a speculative thread.  ... 
doi:10.1145/2355585.2355591 fatcat:qls4y5jz6jhkrj5rlthkbga5wi

Mitosis: A Speculative Multithreaded Processor Based on Precomputation Slices

C. Madriles, C. Garcia-Quinones, J. Sanchez, P. Marcuello, A. Gonzalez, D.M. Tullsen, Hong Wang, J.P. Shen
2008 IEEE Transactions on Parallel and Distributed Systems  
Speculative multithreading increases single-threaded application performance by exploiting thread-level parallelism speculatively, that is, executing code in parallel, even when the compiler or runtime  ...  This paper presents the Mitosis framework, which is a combined hardware-software approach to speculative multithreading, even in the presence of frequent dependences among threads.  ...  A more generic scheme for partitioning the program into speculative threads has been recently presented [17] . This work differs in the way speculative thread inputs are predicted.  ... 
doi:10.1109/tpds.2007.70797 fatcat:qzfb3dwygffmvofhvau6buv5pq

Accurate branch prediction for short threads

Bumyong Choi, Leo Porter, Dean M. Tullsen
2008 ACM SIGOPS Operating Systems Review  
Using a Speculative Multithreaded (SpMT) architecture as an example of a system which generates shorter threads, this work examines techniques to improve branch prediction accuracy when a new thread begins  ...  Thus, modern predictors are almost useless for threads below a certain length.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their helpful insights.  ... 
doi:10.1145/1353535.1346298 fatcat:d6ejoxszkzdgvatavunmxekjay

Accurate branch prediction for short threads

Bumyong Choi, Leo Porter, Dean M. Tullsen
2008 SIGARCH Computer Architecture News  
Using a Speculative Multithreaded (SpMT) architecture as an example of a system which generates shorter threads, this work examines techniques to improve branch prediction accuracy when a new thread begins  ...  Thus, modern predictors are almost useless for threads below a certain length.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their helpful insights.  ... 
doi:10.1145/1353534.1346298 fatcat:lgccezyrcjagjoopui3ytuxawm

Accurate branch prediction for short threads

Bumyong Choi, Leo Porter, Dean M. Tullsen
2008 Proceedings of the 13th international conference on Architectural support for programming languages and operating systems - ASPLOS XIII  
Using a Speculative Multithreaded (SpMT) architecture as an example of a system which generates shorter threads, this work examines techniques to improve branch prediction accuracy when a new thread begins  ...  Thus, modern predictors are almost useless for threads below a certain length.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their helpful insights.  ... 
doi:10.1145/1346281.1346298 dblp:conf/asplos/ChoiPT08 fatcat:2tq33po525dd5ksn7fnaladab4

Accurate branch prediction for short threads

Bumyong Choi, Leo Porter, Dean M. Tullsen
2008 SIGPLAN notices  
Using a Speculative Multithreaded (SpMT) architecture as an example of a system which generates shorter threads, this work examines techniques to improve branch prediction accuracy when a new thread begins  ...  Thus, modern predictors are almost useless for threads below a certain length.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their helpful insights.  ... 
doi:10.1145/1353536.1346298 fatcat:6qp33uprcngfrihhaccddqchiy

P-slice based efficient speculative multithreading

Rakesh Ranjan, Pedro Marcuello, Fernando Latorre, Antonio Gonzalez
2009 2009 International Conference on High Performance Computing (HiPC)  
Also, we propose a scheme called Slice Specialization that reduces the number of instructions in the pre-computation slices by adapting the slice to every single speculative thread spawned.  ...  Speculative multithreading (SpMT) has been proposed in the past to boost performance of irregular applications in multi-core environments.  ...  We compare the effectiveness of our technique with Mitosis [10] which is the state of the art scheme for speculative multithreading using p-slices.  ... 
doi:10.1109/hipc.2009.5433216 dblp:conf/hipc/RanjanMLG09 fatcat:2wshaep3mrbxllprug4s3dhlfu

Threaded multiple path execution

Steven Wallace, Brad Calder, Dean M. Tullsen
1998 SIGARCH Computer Architecture News  
This paper describes the hardware mechanisms needed to enable an SMT processor to efficiently spawn speculative threads for threaded multi-path execution.  ...  This paper presents Threaded Multi-Path Execution (TME), which exploits existing hardware on a Simultaneous Multithreading (SMT) processor to speculatively execute multiple paths of execution.  ...  Acknowledgments We would like to thank Dirk Grunwald, Jack Lo, Susan Eggers, and the anonymous reviewers for providing useful comments. This work was funded in part by NSF CAREER grant No.  ... 
doi:10.1145/279361.279392 fatcat:wsdmrfwnj5f3xo4ud6brrmnbdu

Hardware support for software controlled multithreading

Aqeel Mahesri, Nicholas J. Wang, Sanjay J. Patel
2007 SIGARCH Computer Architecture News  
It divides a single thread of execution into multiple using the master-worker paradigm where some set of master threads execute code that spawns tasks for other, worker theads.  ...  CPU P0 CPU P1 spawn queue communication network L1 L1 L2  ...  Speculative multithreading More recently, a variety of speculative multithreading architectures have been proposed.  ... 
doi:10.1145/1241601.1241606 fatcat:aabcvs7w5vgw3o3bqtmgx3hdpa

Qinling: A Parametric Model in Speculative Multithreading

Yuxiang Li, Yinliang Zhao, Bin Liu
2017 Symmetry  
Speculative multithreading (SpMT) is a thread-level automatic parallelization technique that can accelerate sequential programs, especially for irregular applications that are hard to be parallelized by  ...  Thread partition plays a critical role in SpMT.  ...  Acknowledgments: We thank our colleagues for their collaboration and the present work. We also thank all the reviewers for their specific comments and suggestions.  ... 
doi:10.3390/sym9090180 fatcat:hz5343sctzcydcmkmtsmg2vmru
« Previous Showing results 1 — 15 out of 534 results