Filters








6,175 Hits in 5.8 sec

Page 1463 of Mathematical Reviews Vol. , Issue 2000b [page]

2000 Mathematical Reviews  
“We give two examples of algorithms whose performance is enhanced while retaining a simple modular structure: bounded concurrent timestamping and bounded randomized consensus.  ...  1463 68W Algorithms neous’ picture of a region of shared memory that may be updated by concurrent processes.  ... 

Concurrent Robin Hood Hashing [article]

Robert Kelly, Barak A. Pearlmutter, Phil Maguire
2018 arXiv   pre-print
We compare our data-structures to various other lock-free and concurrent algorithms, as well as a simple hardware transactional variant, and show that our implementation performs better across a number  ...  In this paper we examine the issues involved in adding concurrency to the Robin Hood hash table algorithm.  ...  As CPU utilisation effectively becomes a function of the memory bottleneck [11] , algorithms that use the CPU cache more efficiently can enhance performance for memory bounded tasks.  ... 
arXiv:1809.04339v2 fatcat:3js5jnisljdpxppagj7mp5ieny

Concurrent Robin Hood Hashing

Robert Kelly, Barak A. Pearlmutter, Phil Maguire, Michael Wagner
2018 International Conference on Principles of Distributed Systems  
We compare our data-structures to various other lock-free and concurrent algorithms, as well as a simple hardware transactional variant, and show that our implementation performs better across a number  ...  In this paper we examine the issues involved in adding concurrency to the Robin Hood hash table algorithm.  ...  As CPU utilisation effectively becomes a function of the memory bottleneck [11] , algorithms that use the CPU cache more efficiently can enhance performance for memory bounded tasks.  ... 
doi:10.4230/lipics.opodis.2018.10 dblp:conf/opodis/KellyPM18 fatcat:5vxt7s2wbrckrix77rfm5wakou

Time-based transactional memory with scalable time bases

Torvald Riegel, Christof Fetzer, Pascal Felber
2007 Proceedings of the nineteenth annual ACM symposium on Parallel algorithms and architectures - SPAA '07  
We present a scalable replacement for this global counter and describe how the Lazy Snapshot Algorithm (LSA), which forms the basis for our LSA-STM time-based software transactional memory, has to be changed  ...  Current implementations of time-based transactional memories use a single global clock that is incremented by the commit operation for each update transaction that commits.  ...  Time-based Transactional Memory A transactional memory consists of a set of shared objects that can be accessed concurrently by multiple processors or threads.  ... 
doi:10.1145/1248377.1248415 dblp:conf/spaa/RiegelFF07 fatcat:h5b5fgsvkbaxrcj7i2e6mmycoe

Combinable memory-block transactions

Guy E. Blelloch, Phillip B. Gibbons, S. Harsha Vardhan
2008 Proceedings of the twentieth annual symposium on Parallelism in algorithms and architectures - SPAA '08  
We show how MBTs can be used to design simple and efficient implementations of a variety of protocols and algorithms, including a priority write, a semaphore with a nonblocking P operation, a bounded queue  ...  To gain an understanding of the efficiency that can be gained by using combining, we define a notion of bounded contention and show that all our protocols have bounded contention under arbitrary loads.  ...  Acknowledgments Blelloch and Vardhan were supported in part by NSF Grant CCR-0122581, a gift from Intel and by Microsoft as part of the Carnegie Mellon Center for Computational Thinking.  ... 
doi:10.1145/1378533.1378537 dblp:conf/spaa/BlellochGV08 fatcat:3lnfn755mfhi3l66rgpnojvjdu

Obstruction-Free Algorithms Can Be Practically Wait-Free [chapter]

Faith Ellen Fich, Victor Luchangco, Mark Moir, Nir Shavit
2005 Lecture Notes in Computer Science  
We present a transformation that converts any obstruction-free algorithm into one that is wait-free when analyzed in the unknown-bound semisynchronous model.  ...  Because all practical systems satisfy the assumptions of the unknown-bound model, our result implies that, for all practical purposes, obstruction-free implementations can provide progress guarantees equivalent  ...  Fischer [10] was the first to propose a timing-based mutual exclusion algorithm. He showed that in a known-bound model, there is a simple and efficient algorithm that uses a single shared variable.  ... 
doi:10.1007/11561927_8 fatcat:fdzoxthmwfcqvlzppvy3kzneka

An Efficient Long-Lived Adaptive Collect Algorithm [chapter]

Burkhard Englert
2005 Lecture Notes in Computer Science  
Hence, if the interval contention k rises above K such that K 3 >> N this algorithm becomes inefficient. In this paper, we present a new long-lived, efficient, adaptive collect algorithm.  ...  Moreover, for K such that K 3 ≤ N our algorithm requires only O(N 2 ) shared memory registers. Recently, a number of different adaptive collect algorithms were presented [5, 9, 10, 11, 12] .  ...  Adaptive algorithms have a worst case step complexity that is bounded by a function of the number of concurrently participating, or actually active processors [2] .  ... 
doi:10.1007/11561927_48 fatcat:fgybjg7qjbh23lafy3qdzq6woy

Fast and lock-free concurrent priority queues for multi-thread systems

Håkan Sundell, Philippas Tsigas
2005 Journal of Parallel and Distributed Computing  
We present an efficient and practical lock-free implementation of a concurrent priority queue that is suitable for both fully concurrent (large multi-processor) systems as well as pre-emptive (multi-process  ...  Our algorithm is based on the randomized sequential list structure called Skiplist, and a real-time extension of our algorithm is also described.  ...  In this paper, we present a lock-free algorithm of a concurrent priority queue that is designed for efficient use in both pre-emptive as well as in fully concurrent environments.  ... 
doi:10.1016/j.jpdc.2004.12.005 fatcat:ksg3hmjdonffzgo6aw5jto27ky

Using Shared Memory Abstractions to Design Eager Sequentializations for Weak Memory Models [chapter]

Ermenegildo Tomasco, Truc Lam Nguyen, Bernd Fischer, Salvatore La Torre, Gennaro Parlato
2017 Lecture Notes in Computer Science  
Here we describe a new approach to embed weak memory models within eager sequentializations.  ...  We use our approach to implement a new, efficient BMC-based bug finding tool for multi-threaded C programs under SC, TSO, or PSO based on these SMAs, and show experimentally that it is competitive to existing  ...  of memory models, not just a specific algorithm.  ... 
doi:10.1007/978-3-319-66197-1_12 fatcat:ex3jgnbdkzfhfjlweqmvfu7j6e

Scaling concurrent log-structured data stores

Guy Golan-Gueta, Edward Bortnikov, Eshcar Hillel, Idit Keidar
2015 Proceedings of the Tenth European Conference on Computer Systems - EuroSys '15  
We present cLSM, an algorithm for scalable concurrency in LSM-DS, which exploits multiprocessor-friendly data structures and non-blocking synchronization. cLSM supports a rich API, including consistent  ...  Our algorithm outperforms state of the art LSM-DS implementations, improving throughput by 1.5x to 2.5x.  ...  Our snapshot management algorithm appears in Algorithm 2. Determining the timestamp of a snapshot is a bit subtle.  ... 
doi:10.1145/2741948.2741973 dblp:conf/eurosys/Golan-GuetaBHK15 fatcat:grq5uvi6krbvzfgosrquf7ogdy

A pleasant stroll through the land of infinitely many creatures

Marcos K. Aguilera
2004 ACM SIGACT News  
This paper considers such systems, and gives algorithms that are new and simple (but not necessarily efficient) for common problems.  ...  Abstract Many distributed algorithms are designed for a system with a fixed set of n processes.  ...  A helping mechanism for a simple problem Henceforth we assume that each process has a unique identifier in N, and that memory supports only read and write operations.  ... 
doi:10.1145/992287.992298 fatcat:gnfkqw3nirfvbaulbtmlrg2ivy

Virtual memory transaction management

Michael Stonebraker
1984 ACM SIGOPS Operating Systems Review  
In this paper we examine the consequences of an operating system providing transaction management in an environment where files are bound into a user's address space.  ...  The discussion focuses on inherent limitations in providing concurrency control and crash recovery services in this environment and on hardware extensions needed to overcome these deficiencies.  ...  C2) Concurrency control for B+-tree index pages should have a special case algorithm applied.  ... 
doi:10.1145/850755.850757 fatcat:fhhoz2zdxvd2lofkm3hwpo7r4u

From causal to z-linearizable transactional memory

Torval Riegel, Christof Fetzer, Heiko Sturzrehm, Pascal Felber
2007 Proceedings of the twenty-sixth annual ACM symposium on Principles of distributed computing - PODC '07  
The current generation of time-based transactional memories (TMs) has the advantage of being simple and efficient, and providing strong linearizability semantics.  ...  Linearizability matches well the goal of TM to simplify the design and implementation of concurrent applications.  ...  Conclusion The linearizability semantics supported by timebased transactional memories (TBTMs) are simple to reason about for the developers and allow for efficient and scalable implementations.  ... 
doi:10.1145/1281100.1281162 dblp:conf/podc/RiegelFSF07 fatcat:kvht3lf7xfbcrheru2mybf3muu

Lazy sequentialization for TSO and PSO via shared memory abstractions

Ermenegildo Tomasco, Truc L. Nguyen, Omar Inverso, Bernd Fischer, Salvatore La Torre, Gennaro Parlato
2016 2016 Formal Methods in Computer-Aided Design (FMCAD)  
Lazy sequentialization is one of the most effective approaches for the bounded verification of concurrent programs.  ...  We show experimentally, both on the SV-COMP concurrency benchmarks and a real world instance, that this approach works well in combination with lazy sequentialization on top of bounded model checking.  ...  We set a 10GB memory limit and a 600s timeout for the simple benchmarks and timeout of 14,400s for the safestack example.  ... 
doi:10.1109/fmcad.2016.7886679 dblp:conf/fmcad/TomascoNI0TP16 fatcat:l7ox4a7kprclvnweuuqk7mkshq

Fast General Distributed Transactions with Opacity using Global Time [article]

Alex Shamis, Matthew Renzelmann, Stanko Novakovic, Georgios Chatzopoulos, Anders T. Gjerdrum, Dan Alistarh, Aleksandar Dragojevic, Dushyanth Narayanan, Miguel Castro
2020 arXiv   pre-print
It uses timestamp ordering based on real time with clocks synchronized to within tens of microseconds across a cluster, and a failover protocol to ensure correctness across clock master failures.  ...  Transactions can simplify distributed applications by hiding data distribution, concurrency, and failures from the application developer.  ...  Recall that we assume the timestamping protocol in Algorithm 1, where is a known bound on drift.  ... 
arXiv:2006.14346v1 fatcat:qijkk3p6a5cg5j33yesyrdai44
« Previous Showing results 1 — 15 out of 6,175 results