Filters








79 Hits in 5.6 sec

Speculative lock elision: enabling highly concurrent multithreaded execution

R. Rajwar, J.R. Goodman
Proceedings. 34th ACM/IEEE International Symposium on Microarchitecture. MICRO-34  
We propose Speculative Lock Elision (SLE), a novel micro-architectural technique to remove dynamically unnecessary lock-induced serialization and enable highly concurrent multithreaded execution.  ...  Successful speculative elision is validated and committed without acquiring the lock.  ...  SLE has the following key features: Enables highly concurrent multithreaded execution: Multiple threads can concurrently execute critical sections guarded by the same lock.  ... 
doi:10.1109/micro.2001.991127 dblp:conf/micro/RajwarG01 fatcat:galdfjv3ibf2dfk76bvqnuq3jy

A runtime system for software lock elision

Amitabha Roy, Steven Hand, Tim Harris
2009 Proceedings of the fourth ACM european conference on Computer systems - EuroSys '09  
Execution proceeds optimistically, dynamically detecting conflicts between accesses by concurrent threads.  ...  In this paper we investigate an alternative: a software runtime system that allows threads to speculatively execute lock-based critical sections in parallel.  ...  We introduce a runtime system that allows multiple threads to execute critical sections via software-implemented speculative lock elision [Rajwar 2001 ] (SLE).  ... 
doi:10.1145/1519065.1519094 dblp:conf/eurosys/RoyHH09 fatcat:xpr7yjrgrbg63b5dgkekvsqglm

Transactional execution: toward reliable, high-performance multithreading

R. Rajwar, J. Goodman
2003 IEEE Micro  
TLR uses Speculative Lock Elision (SLE) 2 as an enabling mechanism. In SLE, the hardware dynamically identifies synchronization operations, predicts them as being unnecessary, and elides them.  ...  execute the critical section speculatively.  ...  TLR uses Speculative Lock Elision (SLE) 2 as an enabling mechanism. In SLE, the hardware dynamically identifies synchronization operations, predicts them as being unnecessary, and elides them.  ... 
doi:10.1109/mm.2003.1261395 fatcat:xebtil52yvec3e6ffe6kw3smqe

Grace

Emery D. Berger, Ting Yang, Tongping Liu, Gene Novark
2009 Proceeding of the 24th ACM SIGPLAN conference on Object oriented programming systems languages and applications - OOPSLA 09  
The shift from single to multiple core architectures means that programmers must write concurrent, multithreaded programs in order to increase application performance.  ...  This paper presents Grace, a software-only runtime system that eliminates concurrency errors for a class of multithreaded programs: those based on fork-join parallelism.  ...  While BOP seeks to increase the performance of sequential code by enabling safe, speculative parallelism, Grace provides sequential semantics for concurrently-executing, fork-join based multithreaded programs  ... 
doi:10.1145/1640089.1640096 dblp:conf/oopsla/BergerYLN09 fatcat:jp37gbn3dfd3pih4shdbyjkef4

Grace

Emery D. Berger, Ting Yang, Tongping Liu, Gene Novark
2009 SIGPLAN notices  
The shift from single to multiple core architectures means that programmers must write concurrent, multithreaded programs in order to increase application performance.  ...  This paper presents Grace, a software-only runtime system that eliminates concurrency errors for a class of multithreaded programs: those based on fork-join parallelism.  ...  While BOP seeks to increase the performance of sequential code by enabling safe, speculative parallelism, Grace provides sequential semantics for concurrently-executing, fork-join based multithreaded programs  ... 
doi:10.1145/1639949.1640096 fatcat:rk3ewyzy7zae3l6jqyzuopotra

Transactional lock-free execution of lock-based programs

Ravi Rajwar, James R. Goodman
2002 Tenth international conference on architectural support for programming languages and operating systems on Proceedings of the 10th international conference on architectural support for programming languages and operating systems (ASPLOS-X) - ASPLOS '02  
We seek to improve multithreaded programming trade-offs by providing architectural support for optimistic lock-free execution.  ...  In a lock-free execution, shared objects are never locked when accessed by various threads.  ...  We thank Michael Scott for answering queries regarding MCS behavior, Alaa Alameldeen for discussions regarding multithreaded workload evaluation issues, and Main Kagi and Sanjeev Kumar for the MCS lock  ... 
doi:10.1145/605398.605399 fatcat:ktj45ddok5hubceeyp4f47e7vq

Transactional lock-free execution of lock-based programs

Ravi Rajwar, James R. Goodman
2002 SIGARCH Computer Architecture News  
We seek to improve multithreaded programming trade-offs by providing architectural support for optimistic lock-free execution.  ...  In a lock-free execution, shared objects are never locked when accessed by various threads.  ...  We thank Michael Scott for answering queries regarding MCS behavior, Alaa Alameldeen for discussions regarding multithreaded workload evaluation issues, and Main Kagi and Sanjeev Kumar for the MCS lock  ... 
doi:10.1145/635506.605399 fatcat:xuxd4t7vxzfvbf7pgt3phmghbi

Transactional lock-free execution of lock-based programs

Ravi Rajwar, James R. Goodman
2002 SIGPLAN notices  
We seek to improve multithreaded programming trade-offs by providing architectural support for optimistic lock-free execution.  ...  In a lock-free execution, shared objects are never locked when accessed by various threads.  ...  We thank Michael Scott for answering queries regarding MCS behavior, Alaa Alameldeen for discussions regarding multithreaded workload evaluation issues, and Main Kagi and Sanjeev Kumar for the MCS lock  ... 
doi:10.1145/605432.605399 fatcat:k7j2qiu2m5butopikgfmozuq5i

Transactional lock-free execution of lock-based programs

Ravi Rajwar, James R. Goodman
2002 ACM SIGOPS Operating Systems Review  
We seek to improve multithreaded programming trade-offs by providing architectural support for optimistic lock-free execution.  ...  In a lock-free execution, shared objects are never locked when accessed by various threads.  ...  We thank Michael Scott for answering queries regarding MCS behavior, Alaa Alameldeen for discussions regarding multithreaded workload evaluation issues, and Main Kagi and Sanjeev Kumar for the MCS lock  ... 
doi:10.1145/635508.605399 fatcat:tqajaw76mndoviogkdlctdeqpi

Transactional lock-free execution of lock-based programs

Ravi Rajwar, James R. Goodman
2002 Tenth international conference on architectural support for programming languages and operating systems on Proceedings of the 10th international conference on architectural support for programming languages and operating systems (ASPLOS-X) - ASPLOS '02  
We seek to improve multithreaded programming trade-offs by providing architectural support for optimistic lock-free execution.  ...  In a lock-free execution, shared objects are never locked when accessed by various threads.  ...  We thank Michael Scott for answering queries regarding MCS behavior, Alaa Alameldeen for discussions regarding multithreaded workload evaluation issues, and Main Kagi and Sanjeev Kumar for the MCS lock  ... 
doi:10.1145/605397.605399 dblp:conf/asplos/RajwarG02 fatcat:xyehpp674ffxdolrqmqoox62em

Data-race exceptions have benefits beyond the memory model

Benjamin P. Wood, Luis Ceze, Dan Grossman
2011 Proceedings of the 2011 ACM SIGPLAN Workshop on Memory Systems Performance and Correctness - MSPC '11  
We apply these insights to concurrent garbage collection, optimistic synchronization elision, and best-effort automatic recovery from exceptions due to sequential-consistency-violating races.  ...  Proposals to treat data races as exceptions provide simplified semantics for shared-memory multithreaded programming languages and memory models by guaranteeing that execution remains data-race-free and  ...  Future Work The viability of GC integration with DRE or CRE support merits careful consideration, especially for concurrent copying GC.  ... 
doi:10.1145/1988915.1988923 dblp:conf/pldi/WoodCG11 fatcat:btchrf4qfzewxgu2shahzuvvua

Billion-transistor architectures: there and back again

D. Burger, J.R Goodman
2004 Computer  
23 Speculative Lock Elision allows programmers to include locks that suffer no performance penalty if no lock contention occurs, and the more aggressive Transactional Lock Removal 24 provides lock-free  ...  We (Jim Goodman, along with his colleague, Ravi Rajwar) have been developing hardware support that improves the ease of productive parallel programming by enabling concurrent execution of transactions.  ... 
doi:10.1109/mc.2004.1273999 fatcat:g3pgpbucqfbnbjdjdvxrygjtui

Invyswell

Irina Calciu, Justin Gottschlich, Tatiana Shpeisman, Gilles Pokam, Maurice Herlihy
2014 Proceedings of the 23rd international conference on Parallel architectures and compilation - PACT '14  
After describing Invyswell's design, we show that it outperforms NOrec, a state-of-the-art STM, by 35%, Hybrid NOrec, NOrec's hybrid implementation, by 18%, and Haswell's hardware-only lock elision by  ...  Because of this, a software fallback system must be used in conjunction with Haswell's RTM to ensure transactional programs execute to completion.  ...  Haswell supports hardware lock elision (HLE), which allows an annotated critical section to be first executed speculatively as a hardware transaction, and then, if that transaction fails, to be re-executed  ... 
doi:10.1145/2628071.2628086 dblp:conf/IEEEpact/CalciuGSPH14 fatcat:5ujo43gb2jeypde676evbqkjr4

Hardware tansactional memory support for lightweight dynamic language evolution

Nicholas Riley, Craig Zilles
2006 Companion to the 21st ACM SIGPLAN conference on Object-oriented programming systems, languages, and applications - OOPSLA '06  
With small changes, a runtime can be made HTM-aware to enable parallel execution of Python code and extension modules.  ...  We exploit the semantics of Python execution to evaluate individual bytecodes atomically by default, using nested transactions to emulate programmerspecified locking constructs where possible in existing  ...  In particular, Speculative Lock Elision (SLE) [21] enables concurrent execution of multiple threads holding the same lock as long as their behavior does not conflict.  ... 
doi:10.1145/1176617.1176758 dblp:conf/oopsla/RileyZ06 fatcat:ghn7hmicevdh7plv7yosemiddi

Algorithmic improvements for fast concurrent Cuckoo hashing

Xiaozhou Li, David G. Andersen, Michael Kaminsky, Michael J. Freedman
2014 Proceedings of the Ninth European Conference on Computer Systems - EuroSys '14  
We find that naively allowing concurrent access using a coarse-grained lock on existing data structures reduces overall performance with more threads.  ...  On a 16-core machine, our hash table executes almost 40 million insert and more than 70 million lookup operations per second.  ...  The simplest fallback mechanisms is "lock elision": the program executes a lock-protected region speculatively as a transaction, and only falls back to use normal locking if the transaction does not succeed  ... 
doi:10.1145/2592798.2592820 dblp:conf/eurosys/LiAKF14 fatcat:adodywrjljgbrdceaygagcu7am
« Previous Showing results 1 — 15 out of 79 results