Filters








25,987 Hits in 4.1 sec

A new blocking problem from Java-based schedulers

Peter G. Harrison, Catalina M. Lladó
2003 Performance evaluation (Print)  
´Ã Å µ ½ ¼´Å ص Å Ø Ø Ò Ò Ø ÒØ Ö Ø ÓÒ Ú Ö Ð ÖÓÑ Ø ØÓ × Å Ø Ò Ý Ð × ´Ã µ ´Ã ŵ Å ´Ã ½µ à Š¼´Ã ½µ ´Ã Å µ ½ ¼ × × × ´Ã ŵ Å ´Ã ½µ à Š¼´Ã ½µ ´Ã Å µ ´Ã ŵ Å ´Ã ½µ à Š¼´ · Å ½µ ´½¼μ Ý Ò Ò Ø ×ÙÑÑ Ø ÓÒ Ú Ö  ...  ¢ ¼ È Ã Å´Ã È ½µ ´Å ½µ ´Ã Å È µ Å ½´ ص Ø ´Ã ŵ ´Ã ½µ Å Ø Ã Å ¼´Ã ½µ ´Ã Å µ ¼ È Å ½´ ص ´Ã ŵ ´Ã ½µ Å Ø Ã Å ¼´Ã ½µ ´Ã Å µ ´Å ص ´ µ Ò Ø Ö ×ÙÐØ ÓÐÐÓÛ׺ ¾ Ï ××ÙÑ Ø Ø ÐÐ Ø Å × ÖÚ Ö× Ö ÕÙ ÐÐÝ Ð ÐÝ ØÓ Ó× Ò  ... 
doi:10.1016/s0166-5316(02)00096-2 fatcat:x5ao5zk56fgufkcjziqhigoj7m

An enhanced thread synchronization mechanism for Java

Hsin-Ta Chiao, Shyan-Ming Yuan
2001 Software, Practice & Experience  
In this paper, we first analyze these drawbacks in depth before proceeding to present our own proposal, which is a new monitor-based thread synchronization mechanism that we term EMonitor.  ...  The thread synchronization mechanism of Java is derived from Hoare's monitor concept. In the authors' view, however, it is over simplified and suffers the following four drawbacks.  ...  We are also much obliged to our colleague, Pin-Huang Hsin, who helped solve problems during the coding of the EMonitor's preprocessor.  ... 
doi:10.1002/spe.383 fatcat:fme3k63ojzfs7p2cr7rrbrntiu

Extending JPF to verify distributed systems

Nastaran Shafiei, Peter Mehlitz
2014 Software engineering notes  
It uses a form of partial order reduction (POR) to explore all possible executions of a distributed Java application.  ...  We refer to distributed applications as a collection of communicating processes, regardless of their physical locations and the communication means. Our work targets applications written in Java.  ...  Since the client has not written any data yet, the server blocks and a new state (the state 2) is created from which only the client is enabled.  ... 
doi:10.1145/2557833.2560577 fatcat:52wmm5tuhnbuhf745ayuy4odim

Jcluster: an efficient Java parallel environment on a large-scale heterogeneous cluster

Bao-Yin Zhang, Guang-Wen Yang, Wei-Min Zheng
2006 Concurrency and Computation  
It also shows that the scheduler based on TRS can make any idle node obtain a task from another node with much fewer stealing times on a large-scale cluster.  ...  Performance evaluations show that the scheduler based on TRS can make any idle node obtain a task from another node with much fewer stealing times than random stealing (RS), which is a well-known dynamic  ...  The evaluations show that the task scheduler based on TRS can make any idle node obtain a task from another node with much fewer stealing times on a large-scale cluster.  ... 
doi:10.1002/cpe.986 fatcat:uayqgokolbgf5mi2n52d2jlwky

Lightweight Firm Real-Time Extensions for Low Memory Profile Java Systems [chapter]

Helge Böhme, Ulrich Golze
2004 Lecture Notes in Computer Science  
A three level scheduling scheme using earliest deadline first (EDF), fair-share and round-robin policies is integrated into a Java virtual machine kernel.  ...  Dynamic time-driven firm real-time scheduling is rendered possible by merely using regular Java threads.  ...  If a thread changes from blocked to ready the priority counter is pushed to a high value considering the Java priority and it gets a probability boost. This mechanism doesn't affect deadlines.  ... 
doi:10.1007/978-3-540-30470-8_47 fatcat:2fnsgafznbfvbiwwn6mjqcq3vy

Alma Proposal Tools

Pamela Klaassen, Alan Bidger, Joe Barrett, Andy Biggs, Craige Bevil, Kurowski, Mark Nicol
2018 Zenodo  
Scheduling Blocks?  ...  Same tool -different views. • Rich existing shared domain model represented by the APDM • Thoughtful ongoing involvement Proposed New OT Architecture • Moving from desktop based (thick client  ...  in Python and Java) • Assess performance of new choices for our current resource hot spots (e.g.  ... 
doi:10.5281/zenodo.1292795 fatcat:zr4ir2agwbdw7mayiajb47ol64

Parallel Java: A Unified API for Shared Memory and Cluster Parallel Programming in 100% Java

Alan Kaminsky
2007 2007 IEEE International Parallel and Distributed Processing Symposium  
Parallel Java is a parallel programming API whose goals are (1) to support both shared memory (thread-based) parallel programming and cluster (message-based) parallel programming in a single unified API  ...  a heterogeneous computing environment of single-core CPUs, multi-core CPUs, and clusters thereof.  ...  However, PJ uses multi-threaded blocking I/O ("Old" I/O) to send to and receive from the socket channels, rather than single-threaded selector-based I/O; timing measurements show that blocking I/O gives  ... 
doi:10.1109/ipdps.2007.370421 dblp:conf/ipps/Kaminsky07 fatcat:zimhtmvjirc6bmibmoo3nwxuci

Tasks

Jeffrey Fischer, Rupak Majumdar, Todd Millstein
2007 Proceedings of the 2007 ACM SIGPLAN symposium on Partial evaluation and semantics-based program manipulation - PEPM '07  
A linkable scheduler architecture permits tasks to be used in many different contexts. We have instantiated our model as a backward-compatible extension to Java, called TaskJava.  ...  We illustrate the benefits of our language through a formalization in an extension to Featherweight Java, and through a case study based on an open-source web server.  ...  We provide a direct operational semantics for CTJ, whereby wait calls block until an appropriate event is signaled, as well as a translation relation from CTJ to Featherweight Java, which formalizes the  ... 
doi:10.1145/1244381.1244403 dblp:conf/pepm/FischerMM07 fatcat:smvhictokzenvj34wpnhaahmqa

Model-based schedulability analysis of safety critical hard real-time Java programs

Thomas Bøgholm, Henrik Kragh-Hansen, Petur Olsen, Bent Thomsen, Kim G. Larsen
2008 Proceedings of the 6th international workshop on Java technologies for real-time and embedded systems - JTRES '08  
The approach is based on a translation of programs, written in the Safety Critical Java profile introduced in [21] for the Java Optimized Processor [18] , to timed automata models verifiable by the Uppaal  ...  In this paper, we present a novel approach to schedulability analysis of Safety Critical Hard Real-Time Java programs.  ...  A web based version of the SARTS tools has been developed, and is available at http://sarts.boegholm.dk/. ACKNOWLEDGMENT We would like to thank Anders P.  ... 
doi:10.1145/1434790.1434807 dblp:conf/jtres/BogholmKOTL08 fatcat:s62tfvlksvhezkxhgo7oieg7dq

Linguistic symbiosis between actors and threads

Tom Van Cutsem, Stijn Mostinckx, Wolfgang De Meuter
2007 Proceedings of the 2007 international conference on Dynamic languages in conjunction with the 15th International Smalltalk Joint Conference 2007 - ICDL '07  
We describe a linguistic symbiosis between AmbientTalk, a flexible, domain-specific language for writing distributed programs and Java, a conventional object-oriented language.  ...  AmbientTalk employs an inherently event-driven model based on actors, while conventional object-oriented languages employ a concurrency model based on threads.  ...  AmbientTalk's futures are based on the E language's non-blocking futures [8] : an event loop is never allowed to block, waiting for a future to become fulfilled.  ... 
doi:10.1145/1352678.1352693 dblp:conf/smalltalk/CutsemMM07 fatcat:eeneweskcndnfmut47h2o7u5gi

How to solve the Santa Claus problem

MORDECHAI BEN‐ARI
1998 Concurrency Practice and Experience  
John Trono published a new exercise in concurrent programming-the Santa Claus problem-and provided a solution based on semaphores [12] .  ...  His solution is incorrect because it assumes that a process released from waiting on a semaphore will necessarily be scheduled for execution.  ...  INTRODUCTION John Trono published a new exercise in concurrent programming [12] with the aim of providing educators with a more challenging problem than simple mutual exclusion: a problem with three  ... 
doi:10.1002/(sici)1096-9128(199805)10:6<485::aid-cpe329>3.3.co;2-u fatcat:akl4vpmborbg7dutahxhi5s2i4

How to solve the Santa Claus problem

MORDECHAI BEN-ARI
1998 Concurrency Practice and Experience  
John Trono published a new exercise in concurrent programming-the Santa Claus problem-and provided a solution based on semaphores [12] .  ...  His solution is incorrect because it assumes that a process released from waiting on a semaphore will necessarily be scheduled for execution.  ...  INTRODUCTION John Trono published a new exercise in concurrent programming [12] with the aim of providing educators with a more challenging problem than simple mutual exclusion: a problem with three  ... 
doi:10.1002/(sici)1096-9128(199805)10:6<485::aid-cpe329>3.0.co;2-2 fatcat:bdsq57yl7zdk7hiscypnheekiq

Experiences in Adopting Real-Time Java for Flight-Like Software [chapter]

Edward G. Benowitz, Albert F. Niessner
2003 Lecture Notes in Computer Science  
Utilizing a real mission design, this work leverages the original flight code from NASA's Deep Space l(DSl), which flew in 1998.  ...  Thus, realtime client code can be run on a standard Java virtual machine, allowing the code to be debugged in a graphical development environment on a desktop PC at the cost of decreased real-time performance  ...  Since the specification is a new technology, there are few examples of real-time Java being applied to non-trivial problems.  ... 
doi:10.1007/978-3-540-39962-9_54 fatcat:24qoazdwx5cybpqu6fkfvwekmu

Dynamic Load-Balancing and High Performance Communication in Jcluster

Bao-Yin Zhang, Ze-Yao Mo, Guang-Wen Yang, Wei-Min Zheng
2007 2007 IEEE International Parallel and Distributed Processing Symposium  
For the efficient loadbalancing, we implement a task scheduler based on a Transitive Random Stealing algorithm, which improves the Random Stealing, a well-known load-balancing algorithm.  ...  The experiment results show that the scheduler performs efficiently, especially for a large-scale cluster.  ...  And it shows that the scheduler based on TRS can make any idle node obtain a task from another node with much fewer stealing times on a large-scale cluster.  ... 
doi:10.1109/ipdps.2007.370417 dblp:conf/ipps/ZhangMYZ07 fatcat:id4vxflh7fhw5mg2siqcprhhjy

Runtime support for scalable programming in Java

Sang Boem Lim, Hanku Lee, Bryan Carpenter, Geoffrey Fox
2007 Journal of Supercomputing  
Our HPJava is based around a small set of language extensions designed to support parallel computation with distributed arrays, plus a set of communication libraries.  ...  Hence, Java looks like a very promising alternative for the future. The paper will discuss in detail a particular environment called HPJava.  ...  Some of the range classes allow ghost extensions to support stencil-based computations. A second new control construct, overall, implements a distributed parallel loop.  ... 
doi:10.1007/s11227-007-0125-5 fatcat:nlmxxwvftvforg7rumfnyu5vve
« Previous Showing results 1 — 15 out of 25,987 results