Filters








70,189 Hits in 4.3 sec

Efficient dependency tracking for relevant events in concurrent systems

Anurag Agarwal, Vijay K. Garg
2006 Distributed computing  
In a concurrent system with N processes, vector clocks of size N are used for tracking dependencies between the events. Using vectors of size N leads to scalability problems.  ...  We present a class of logical clock algorithms, called chain clock, for tracking dependencies between relevant events based on generalizing a process to any chain in the computation poset.  ...  track relevant events with fewer than N components in many cases.  ... 
doi:10.1007/s00446-006-0004-y fatcat:xpys4wwuh5aotlmm3uv4j3a62y

Brief announcement

Nuno Preguiça, Carlos Bauqero, Paulo Sérgio Almeida, Victor Fonte, Ricardo Gonçalves
2012 Proceedings of the 2012 ACM symposium on Principles of distributed computing - PODC '12  
Version vectors (VV) are used pervasively to track dependencies between replica versions in multi-version distributed storage systems.  ...  In these systems, VV tend to have a dual functionality: identify a version and encode causal dependencies.  ...  We would like to thank Doug Terry for his comments on previous versions of this work.  ... 
doi:10.1145/2332432.2332497 dblp:conf/podc/PreguicaBAFG12 fatcat:k3ys7h6cnjeutjnriuvuaij2ma

Predictive runtime verification of multi-processor SoCs in SystemC

Alper Sen, Vinit Ogale, Magdy S. Abadir
2008 Proceedings of the 45th annual conference on Design automation - DAC '08  
Concurrent interaction of multi-processor systems result in errors which are difficult to find.  ...  We present a novel simulation-based technique for SystemC that preserves and exploits concurrency information.  ...  Potential error detection is possible in our approach due to preserving and exploiting concurrency information by tracking dependencies during the execution of a system.  ... 
doi:10.1145/1391469.1391708 dblp:conf/dac/SenOA08 fatcat:kdkyxzioondk7g35gy4nu6magu

Concurrent Design of Embedded Control Software

Marcel Groothuis, Raymond Frijns, Jeroen Voeten, Jan Broenink
2009 Electronic Communications of the EASST  
designed embedded system software parts in an efficient way.  ...  In this paper, we present a case study on the development of the embedded control software for a real-world mechatronic system in order to evaluate how we can integrate concurrent and largely independent  ...  Our methodology aims at a concurrent design process with minimal but sufficient information-exchange between the concurrent tracks (i.e. maximizing concurrency efficiency), while still relying on a predictable  ... 
doi:10.14279/tuj.eceasst.21.284 dblp:journals/eceasst/GroothuisFVB09 fatcat:b2d762jtlnewlnkxpny4ivxviq

k-Dependency Vectors: a scalable causality-tracking protocol

R. Baldoni, G. Melideo
2003 Eleventh Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2003. Proceedings.  
In this paper we present a scalable causality-tracking protocol, called -Dependency Vectors, which piggybacks on each application message a constant number of integers (with Ò).  ...  In the paper, several selection strategies are proposed and evaluated with respect to the detection delay experienced by the checker.  ...  In the paper we present -Dependency Vectors, a scalable protocol for causality-tracking which associates each event with a dependency vector -Î of size Ò.  ... 
doi:10.1109/empdp.2003.1183592 dblp:conf/pdp/BaldoniM03 fatcat:nwvwzs3a35gency46hbrashr7y

Using distributed consistent branching for efficient reconciliation of mobile workspaces

Felix Hupfeld, Minor Gordon
2006 2006 International Conference on Collaborative Computing: Networking, Applications and Worksharing  
Simulation results validate our design and show that it compares favorably in the relevant metrics with version vectorbased designs.  ...  When conflict handling requires manual intervention by the user, the system must not require that conflicts are resolved immediatley after their detection.  ...  ACKNOWLEDGEMENT This work was supported in part by the EU IST program in context of the XtreemOS project (contract FP6-033576).  ... 
doi:10.1109/colcom.2006.361837 dblp:conf/colcom/HupfeldG06a fatcat:4ejcvkli2vfynbnp6tv3wm2r44

Using distributed consistent branching for efficient reconciliation of mobile workspaces

Felix Hupfeld, Minor Gordon
2006 2006 International Conference on Collaborative Computing: Networking, Applications and Worksharing  
Simulation results validate our design and show that it compares favorably in the relevant metrics with version vectorbased designs.  ...  When conflict handling requires manual intervention by the user, the system must not require that conflicts are resolved immediatley after their detection.  ...  ACKNOWLEDGEMENT This work was supported in part by the EU IST program in context of the XtreemOS project (contract FP6-033576).  ... 
doi:10.1109/colcom.2006.361860 dblp:conf/colcom/HupfeldG06 fatcat:k3knnrdtlbfnng62w7f47ctsl4

First experiences with a parallel architecture testbed in the LHCb trigger system

S Gallorini, D Lucchesi, A Gianelle, S Amerio, M Corvo
2017 Journal of Physics, Conference Series  
During normal data taking, real events have been sent to the node and processed by GPU-based and CPU-based tracking algorithms.  ...  In this note we will discuss the application of new technologies, such as GPU cards, in the current LHCb trigger system.  ...  Acknowledgements We would like to thank the Online and HLT teams (Niko Neufeld, Markus Frank, Rainer Schwemmer, Mohammed Daoudi, Roel Aaji) for helping us in the installation of the testbed in the monitoring  ... 
doi:10.1088/1742-6596/898/3/032029 fatcat:ewkwcnwlqnaojcuhs6zfawxrmy

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events [article]

Giuseppe Cerati, Peter Elmer, Brian Gravelle, Matti Kortelainen, Vyacheslav Krutelyov, Steven Lantz, Matthieu Lefebvre, Mario Masciovecchio, Kevin McDermott, Boyana Norris, Allison Reinsvold Hall, Daniel Riley, Matevz Tadel, Peter Wittich (+2 others)
2018 arXiv   pre-print
Starting in 2014, we have been developing Kalman-filter-based methods for track finding and fitting adapted for many-core SIMD processors that are becoming dominant in high-performance systems.  ...  The High-Luminosity Large Hadron Collider at CERN will be characterized by greater pileup of events and higher occupancy, making the track reconstruction even more computationally demanding.  ...  support for multiple concurrent events in flight in mkFit as well.  ... 
arXiv:1811.04141v1 fatcat:ibrvwwzy4vdefex6ht2l5qm2ca

Tracking unbounded Topic Streams

Dominik Wurzer, Victor Lavrenko, Miles Osborne
2015 Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)  
for nearest neighbour search on streams.  ...  When tracking 4.4 million topics against 52 million documents in constant time and space, we demonstrate that counter to expectations, simple single-pass clustering can outperform locality sensitive hashing  ...  from an exact tracking system • We show how single-pass clustering can outperform locality sensitive hashing in terms of effectiveness and efficiency for identifying nearest neighbours in a stream • We  ... 
doi:10.3115/v1/p15-1170 dblp:conf/acl/WurzerLO15 fatcat:mkvqkimmfvfvtlwasoveptvk6m

AESOP: Expressing Concurrency in High-Performance System Software

Dries Kimpe, Philip Carns, Kevin Harms, Justin M. Wozniak, Samuel Lang, Robert Ross
2012 2012 IEEE Seventh International Conference on Networking, Architecture, and Storage  
Such system software services must manage highly concurrent requests, interact with a wide range of resources, and scale well in order to be successful.  ...  Unfortunately, no single programming model for distributed system software currently offers optimal performance and productivity for all these tasks.  ...  In the asynchronous cancellation mode, depending on the system, thread execution might be halted at any point, making it hard for the programmer to properly track cleanup state.  ... 
doi:10.1109/nas.2012.41 dblp:conf/nas/KimpeCHWLR12 fatcat:zzta4rqrajck3pbdlpfgkxo2ba

Visual tracking using Sensor Networks

Paolo Pagano, Francesco Piga, Giuseppe Lipari, Yao Liang
2009 Proceedings of the Second International ICST Conference on Simulation Tools and Techniques  
A set of results is carried out estimating the performances of the Visual Tracking system in two contexts (those of a monitored junction in an airport taxiway and in a parking area) very different for  ...  The performances of the deployed system (hereby felt as a black box) must be tracked against the reliability and timeliness offered in message delivery.  ...  In a real-time perspective, the system overhead for event detection is very relevant.  ... 
doi:10.4108/icst.simutools2009.5575 dblp:conf/simutools/PaganoPLL09 fatcat:jakbj43st5bdheig6dfw7bsp6m

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events

Giuseppe Cerati, Peter Elmer, Brian Gravelle, Matti Kortelainen, Vyacheslav Krutelyov, Steven Lantz, Matthieu Lefebvre, Mario Masciovecchio, Kevin McDermott, Boyana Norris, Allison Reinsvold Hall, Daniel Riley (+9 others)
2019 EPJ Web of Conferences  
Starting in 2014, we have been developing Kalman-filter-based methods for track finding and fitting adapted for many-core SIMD processors that are becoming dominant in high-performance systems.  ...  The High-Luminosity Large Hadron Collider at CERN will be characterized by greater pileup of events and higher occupancy, making the track reconstruction even more computationally demanding.  ...  support for multiple concurrent events in flight in mkFit as well [8] .  ... 
doi:10.1051/epjconf/201921402002 fatcat:xynaox3qxjgv5jq7ghtolsle2a

TransJ: An Abstract Independent-Framework for Weaving Crosscutting Concern into Distributed Transactions

Anas M. R. AlSobeh, Stephen W. Clyde
2016 Computer Technology and Application  
To address these problems, we propose an extension to AspectJ framework, called TransJ, that allows developers to define pointcuts in terms of transaction abstractions and that automatically keeps track  ...  concerns in modular, cohesive and loosely coupled transaction-aware aspects.  ...  The aspects can apply for a transaction and keep track of the multiple concurrent transactions by maintaining a collection of contexts.  ... 
doi:10.17265/1934-7332/2016.04.001 fatcat:lmlosh7lz5c4dhx2qj4wnqmw3a

Tracking immediate predecessors in distributed computations

Emmanuelle Anceaume, Jean-Michel Hélary, Michel Raynal
2002 Proceedings of the fourteenth annual ACM symposium on Parallel algorithms and architectures - SPAA '02  
An important causality-related distributed computing problem, that we call the Immediate Predecessors Tracking (IPT) problem, consists in associating with each relevant event, on the fly and without using  ...  additional control messages, the set of relevant events that are its immediate predecessors in the partial order.  ...  Solving this problem requires tracking causality, hence using vector clocks. Previous works have addressed the efficient implementation of vector clocks to track causal dependence on relevant events.  ... 
doi:10.1145/564870.564907 dblp:conf/spaa/AnceaumeHR02 fatcat:3wocuisjqbaurgzbv647v5olvu
« Previous Showing results 1 — 15 out of 70,189 results