Filters








557 Hits in 2.6 sec

Nonblocking collectives for scalable Java communications

Sabela Ramos, Guillermo L. Taboada, Roberto R. Expósito, Juan Touriño
2014 Concurrency and Computation  
Nonblocking collectives aim to exploit the overlapping between computation and communication for collective operations to increase scalability of message passing codes, as it has been done for nonblocking  ...  This paper presents a Java implementation of the recently published MPI 3.0 nonblocking message passing collectives in order to analyze and assess the feasibility of taking advantage of these operations  ...  NONBLOCKING COLLECTIVE LIBRARY FOR MESSAGE PASSING IN JAVA The benefits of nonblocking collectives have been extensively studied for distributed memory systems and communication across the network.  ... 
doi:10.1002/cpe.3279 fatcat:kkze2rl2zfaebgrjivzr7yxxri

F-MPJ: scalable Java message-passing communications on parallel systems

Guillermo L. Taboada, Juan Touriño, Ramón Doallo
2009 Journal of Supercomputing  
This paper presents F-MPJ (Fast MPJ), a scalable and efficient Message-Passing in Java (MPJ) communication middleware for parallel computing.  ...  highperformance networks through the use of our high-performance Java sockets implementation (named JFS, Java Fast Sockets); (3) avoiding the use of communication buffers; and (4) optimizing MPJ collective  ...  We gratefully thank CESGA (Galician Supercomputing Center, Santiago de Compostela, Spain) for providing access to the Finis Terrae supercomputer.  ... 
doi:10.1007/s11227-009-0270-0 fatcat:2uddtgfvund3nhk2iuyimyouxi

Device level communication libraries for high-performance computing in Java

Guillermo L. Taboada, Juan Touriño, Ramón Doallo, Aamir Shafi, Mark Baker, Bryan Carpenter
2011 Concurrency and Computation  
essential to obtain scalability in Java for HPC.  ...  Its public release currently contains three communication devices: the first one is built using the Java New Input/Output (NIO) package for the TCP/IP; the second one is specifically designed for the Myrinet  ...  We gratefully thank CESGA (Galicia Supercomputing Center, Santiago de Compostela, Spain) for providing access to the Finis Terrae supercomputer.  ... 
doi:10.1002/cpe.1777 fatcat:hf77nzvkzzdftja475uaxrxmmu

Scalable synchronous queues

William N. Scherer, Doug Lea, Michael L. Scott
2009 Communications of the ACM  
We present two new nonblocking and contention-free implementations of synchronous queues, concurrent transfer channels in which producers wait for consumers just as consumers wait for producers.  ...  Our synchronous queues have been adopted for inclusion in Java 6.  ...  Acknowledgments We are grateful to Dave Dice, Brian Goetz, David Holmes, Mark Moir, Bill Pugh, and the anonymous referees for giving feedback that significantly improved the presentation of this paper.  ... 
doi:10.1145/1506409.1506431 fatcat:l3oc62fdv5eyxo2f325i6x3qqu

Design of efficient Java message-passing collectives on multi-core clusters

Guillermo L. Taboada, Sabela Ramos, Juan Touriño, Ramón Doallo
2010 Journal of Supercomputing  
This paper presents a scalable and efficient Message-Passing in Java (MPJ) collective communication library for parallel computing on multi-core architectures.  ...  Thus, this paper presents a more scalable collectives communication library.  ...  We gratefully thank CESGA (Galicia Supercomputing Center, Santiago de Compostela, Spain) for providing access to the Finis Terrae supercomputer.  ... 
doi:10.1007/s11227-010-0464-5 fatcat:tjl7kfhidreunirkac2xohvsk4

Distributed Nonblocking Commit Protocols for Many-Party Cross-Blockchain Transactions [article]

Xinying Wang, Olamide Timothy Tawose, Feng Yan, Dongfang Zhao
2020 arXiv   pre-print
Existing techniques either work only for two blockchains or requires a centralized component to govern the cross-blockchain transaction execution, neither of which would meet the scalability requirement  ...  (linear scalability on up to 64-blockchain transactions).  ...  Figure 3 : 3 Blocking 2PC vs. nonblocking CBT. at: https://github.com/hpdic/cbt. The source code is written with Java of JDK 1.7. The code base comprises about 5,190 lines of code.  ... 
arXiv:2001.01174v1 fatcat:6vha5dpzinfmvpl5cm542mr3qu

Avoiding a middleware muddle

J. Thompson
1997 IEEE Software  
Transparent to the user, the Profiler collects data from both local and remote user locations, then automatically records and transports it to the vendor site, where it is compiled for presentation.  ...  to be asynchronous and nonblocking.  ... 
doi:10.1109/52.636685 fatcat:bd5q26gulbcqnbdbzwakohzot4

Towards efficient shared memory communications in MPJ express

Aamir Shafi, Jawad Manzoor
2009 2009 IEEE International Symposium on Parallel & Distributed Processing  
In this paper, we present two shared memory implementations meant for providing efficient communication of multi-core and SMP clusters.  ...  Each Java thread represents an MPI level OS process and communication between these threads is achieved using shared data structures.  ...  We would also like to thank Mr Kamran Hameed, NUST for configuring and benchmarking the mpiJava software on our local cluster.  ... 
doi:10.1109/ipdps.2009.5161083 dblp:conf/ipps/ShafiM09 fatcat:l5t6rxayt5erhg2g5tievlff2q

PCJ Java library as a solution to integrate HPC, Big Data and Artificial Intelligence workloads

Marek Nowicki, Łukasz Górski, Piotr Bała
2021 Journal of Big Data  
We have developed the Parallel Computing in Java (PCJ) library, a tool for scalable high-performance computing and Big Data processing in Java.  ...  The performance results are presented for different classes of applications including traditional computational intensive (HPC) workloads (e.g. stencil), as well as communication-intensive algorithms such  ...  We acknowledge Interdisciplinary Centre for Mathematical and Computational Modelling (ICM) University of Warsaw for providing computational resources under grants no. GB65-15, GA69-19.  ... 
doi:10.1186/s40537-021-00454-6 doaj:f7366bb69e344936853189fa95deea9e fatcat:u2nisnxoufd23om5ttfi6igyc4

High Performance Message-passing InfiniBand Communication Device for Java HPC

Omar Khan, Mohsan Jameel, Aamir Shafi
2014 Procedia Computer Science  
Java message passing libraries have predominantly provided portable communication devices, rather than focus on developing low level devices for communication on high speed networks.  ...  If Java is to become a language of choice for HPC community, the tremendous improvement in its computational performance needs to be augmented with high communication performance on these new network architectures  ...  It obtains performance similar to MPI libraries because of efficient and scalable communication protocol implementations, topology aware collective primitives, zero-copy protocols to avoid buffering overheads  ... 
doi:10.1016/j.procs.2014.05.133 fatcat:zqcht2axs5dazlhhwcutpzn3vq

FastMPJ: a scalable and efficient Java message-passing library

Roberto R. Expósito, Sabela Ramos, Guillermo L. Taboada, Juan Touriño, Ramón Doallo
2014 Cluster Computing  
The performance and scalability of communications are key for High Performance Computing (HPC) applications in the current multi-core era.  ...  advantage of high-speed cluster networks (e.g., InfiniBand) to provide low-latency and high bandwidth communications; (3) including a scalable collective library with topology aware primitives, automatically  ...  We also gratefully thank the Advanced School for Computing and Imaging (ASCI) and the Vrije University Amsterdam for providing access to the DAS-4 cluster.  ... 
doi:10.1007/s10586-014-0345-4 fatcat:jxpqqaj3frbrfendjhmims72ty

HIGH PRODUCTION OF JAVA SOCKETS (HPJS) FOR HELTH CLOUDS IN SCIENCE

Vasyl Melnyk, Oksana Zhyharevych, Kateryna Melnyk
2015 Vìsnik Nacìonalʹnogo Avìacìjnogo Unìversitetu  
HPJS effectively uses the Java socket realization for high-performance inner-process communications.  ...  Computer clouds are using in health science for its data collections, manipulations and providing security needs in communications to exchange.  ...  It uses the Java-NIO implementation for high-performance applications and network high-scalability applications.  ... 
doi:10.18372/2306-1472.64.9041 fatcat:qtauzwognvhffh4nyzaygohe7u

Titanium Performance and Potential: An NPB Experimental Study [chapter]

Kaushik Datta, Dan Bonachea, Katherine Yelick
2006 Lecture Notes in Computer Science  
Titanium is an explicitly parallel dialect of Java TM designed for high-performance scientific programming.  ...  In addition, the use of Java as a base language provides support for strong typing, user-defined classes, inheritance, and dynamic memory management.  ...  support for synchronization and collective communication.  ... 
doi:10.1007/978-3-540-69330-7_14 fatcat:zwatvttukjcu7csidw3qa2rlzu

Scalable Wireless Ad Hoc Network Simulation [chapter]

Rimon Barr, Robbert van Renesse, Zygmunt Haas
2005 Handbook on Theoretical and Algorithmic Aspects of Sensor, Ad Hoc Wireless, and Peer-to-Peer Networks  
Ongoing research into dynamic, self-organizing, multihop wireless networks, called ad hoc networks, promises to improve the efficiency and coverage of wireless communication.  ...  For example, routing and transmitting packets efficiently over ad hoc networks becomes difficult as they grow in size.  ...  To conclude, we hope that the scalability, performance, and flexibility of SWANS, combined with the popularity of the Java language, will facilitate its broader adoption within the network simulation community  ... 
doi:10.1201/9780203323687.ch19 fatcat:ukjt5f6upren7biuan6tpjjpmy

Scalable software defined network controllers

Andreas Voellmy, Junchang Wang
2012 Proceedings of the ACM SIGCOMM 2012 conference on Applications, technologies, architectures, and protocols for computer communication - SIGCOMM '12  
The simplicity of a logical centralized controller, however, can come at the cost of control-plane scalability.  ...  McNettle can serve up to 5000 switches using a single controller with 46 cores, achieving throughput of over 14 million flows per second, near-linear scaling up to 46 cores, and latency under 200 μs for  ...  PROGRAMMING MODEL McNettle programs consist of a collection of message handlers, one for each switch in the network.  ... 
doi:10.1145/2342356.2342414 dblp:conf/sigcomm/VoellmyW12 fatcat:f4gcsfs3lrfddcbrt6u3zbzgva
« Previous Showing results 1 — 15 out of 557 results