Filters








26,069 Hits in 8.4 sec

Limitations of Cycle Stealing for Parallel Processing on a Network of Homogeneous Workstations

Scott T. Leutenegger, Xian-He Sun
1997 Journal of Parallel and Distributed Computing  
We propose that task ratio is a useful metric for determining how a parallel applications should be partitioned and scheduled in order to make efficient use of a nondedicated distributed system.  ...  It forms a foundation for task partitioning and scheduling in a nondedicated network environment.  ...  We assume the parallel application considered belongs to the class of programs that can run efficiently in a dedicated distributed computing environment.  ... 
doi:10.1006/jpdc.1997.1341 fatcat:zeixolzjyfdq7cmo4mlcuxhaeq

Knowledge engineering for automatic parallel performance diagnosis

L. Li, A. D. Malony
2007 Concurrency and Computation  
General metrics for evaluating performance includes execution time, parallel overhead, speedup, efficiency, and cost [11] .  ...  We present a systematic approach to generating performance knowledge for automatically diagnosing parallel programs.  ...  ACKNOWLEDGEMENTS We would like to thank Professor Ginnie Lo for reviewing and valuable comments on earlier versions of this paper.  ... 
doi:10.1002/cpe.1127 fatcat:nekxfrlnofgdloup3pyyywcgpa

INTEGRATED EXTREMAL OPTIMIZATION AND RANDOM FOREST BASED SCHEDULING FOR CLOUD COMPUTING ENVIRONMENT

2017 International Journal of Advanced Research in Computer Science  
Basically it is a parallel and distributed system containing various powerfully interconnected virtualized machines. Virtual machines are assigned over system to give administrations to users.  ...  Cloud computing is the most recent continuation of parallel computing, distributed computing and grid computing.  ...  metrics , value of this function is in the range [0, 1], i.e., it is equal to 0 when there is no migration, when all tasks have to be migrated , migration(Y) = 1, otherwise 0 ≤ migration(Y) ≤ 1 [3]: Where  ... 
doi:10.26483/ijarcs.v8i7.4150 fatcat:lgeyybyvlfafhbqqaoaa5t2b5e

OS Support for Thread Migration and Distribution in the Fully Heterogeneous Datacenter

Pierre Olivier, Sang-Hoon Kim, Binoy Ravindran
2017 Proceedings of the 16th Workshop on Hot Topics in Operating Systems - HotOS '17  
pro les and we show that signi cant potential bene ts for both metrics can be expected, should these machines be able to cooperate in the processing of datacenter, multiprogrammed workloads.  ...  image, a distributed shared memory system, and a heterogeneous scheduler.  ...  For a given workload, di erent types of machines will exhibit di erent performance and power consumption behaviors due to the workload characteristics, de ned according to multiple metrics: potential parallelism  ... 
doi:10.1145/3102980.3103009 dblp:conf/hotos/OlivierKR17 fatcat:hhrwtk7tujhwtgxgz3zaae5bja

Scheduling Multiple Multithreaded Applications on Asymmetric and Symmetric Chip Multiprocessors

T Y Morad, A Kolodny, U C Weiser
2010 2010 3rd International Symposium on Parallel Architectures, Algorithms and Programming  
This paper evaluates new techniques to improve performance, fairness and jitter of workloads consisting of multiple multithreaded applications running on Chip MultiProcessors (CMP).  ...  Multithreaded applications contain serial phases (single thread) and parallel phases (many threads).  ...  They would also like to thank Andrey Gelman and Niv Aibester for their help in setting up the emulation environment.  ... 
doi:10.1109/paap.2010.50 dblp:conf/paap/MoradKW10 fatcat:h46whmoygvexfnq5zg3cgvws3e

Model-based, memory-centric performance and power optimization on NUMA multiprocessors

ChunYi Su, Dong Li, Dimitrios S. Nikolopoulos, Kirk W. Cameron, Bronis R. de Supinski, Edgar A. Leon
2012 2012 IEEE International Symposium on Workload Characterization (IISWC)  
Using an ANN in conjunction with critical path analysis, DyNUMA autonomously optimizes programs for performance or energy-efficiency metrics.  ...  DyNUMA uses simple, memorycentric, performance and energy models with non-linear terms to capture the complex and interacting effects of system layout, program concurrency, data placement, and memory controller  ...  Government retains a non-exclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes.  ... 
doi:10.1109/iiswc.2012.6402921 dblp:conf/iiswc/SuLNCSL12 fatcat:mdsdggdgq5c57oc3nfchrxkzke

Distributed Data Stream Processing and Edge Computing: A Survey on Resource Elasticity and Future Directions [article]

Marcos Dias de Assuncao, Alexandre da Silva Veith, Rajkumar Buyya
2017 arXiv   pre-print
Elasticity becomes even more challenging in highly distributed environments comprising edge and cloud computing resources.  ...  Resource elasticity allows for an application or service to scale out/in according to fluctuating demands.  ...  This work has been carried out in the scope of a joint project between the French National Center for Scientific Research (CNRS) and the University of Melbourne.  ... 
arXiv:1709.01363v2 fatcat:ajven75pjrgqhkpmi2d3pxs5pu

Distributed data stream processing and edge computing: A survey on resource elasticity and future directions

Marcos Dias de Assunção, Alexandre da Silva Veith, Rajkumar Buyya
2018 Journal of Network and Computer Applications  
Elasticity becomes even more challenging in highly distributed environments comprising edge and cloud computing resources.  ...  Resource elasticity allows for an application or service to scale out/in according to fluctuating demands.  ...  This work has been carried out in the scope of a joint project between the French National Center for Scientific Research (CNRS) and the University of Melbourne.  ... 
doi:10.1016/j.jnca.2017.12.001 fatcat:twmpqzkb3nco3a7nwyhloe5qvu

Analysis of Requirements for Virtual Machine Migration in Dynamic Clouds

Stelios Sotiriadis, Nik Bessis, Pawel Gepner, Nicolas Markatos
2013 2013 IEEE 12th International Symposium on Parallel and Distributed Computing  
Highly dynamic environments like clouds by nature cause a high degree of unpredictability of resource utilization and performance.  ...  Failures, latencies and heterogeneity should always be the main concern for affecting the scheduling decisions in distributed infrastructures.  ...  Those solutions are considered as host driven migration and benchmarks shows that performance was generally on parallel with VMWare.  ... 
doi:10.1109/ispdc.2013.24 dblp:conf/ispdc/SotiriadisBGM13 fatcat:owcc55tyhrertcqpmxjkwzmu4i

Dynamic Load Balancing Based on Applications Global States Monitoring

Eryk Laskowski, Marek Tudruj, Richard Olejnik, Damian Kopanski
2013 2013 IEEE 12th International Symposium on Parallel and Distributed Computing  
Methods for designing processor load balancing control based on a system of program and system properties metrics and computational data migration between application executive processes is presented and  ...  The paper presents how to use a special novel distributed program design framework with evolved global control mechanisms to assure processor load balancing during execution of application programs.  ...  In our strategy, we propose to avoid such strategy and to replace it by workload distribution control, however, allowing for real load migration if unavoidable by other methods.  ... 
doi:10.1109/ispdc.2013.11 dblp:conf/ispdc/LaskowskiTOK13 fatcat:ievg6t3fgvdiznbyifcymljei4

CHAMELEON: Reactive load balancing for hybrid MPI+openMP task-parallel applications

Jannis Klinkenberg, Philipp Samfass, Michael Bader, Christian Terboven, Matthias S. Müller
2019 Journal of Parallel and Distributed Computing  
Aiming to tackle these challenges in search for a general solution, we present a novel library for fine-granular task-based reactive load balancing in distributed memory based on MPI and OpenMP.  ...  Many applications in high performance computing are designed based on underlying performance and execution models.  ...  Further, we like to thank Marcus Wagner and Sascha Bücken for the excellent technical support and administrative help during our evaluation phase.  ... 
doi:10.1016/j.jpdc.2019.12.005 fatcat:ffhc5ppnqfdttlbtzwmrhx3sfm

Job Superscheduler Architecture and Performance in Computational Grid Environments

Hongzhang Shan, Leonid Oliker, Rupak Biswas
2003 Proceedings of the 2003 ACM/IEEE conference on Supercomputing - SC '03  
In this paper, we propose a novel grid superscheduler architecture and three distributed job migration algorithms.  ...  Several key metrics demonstrate that substantial performance gains can be achieved via smart superscheduling in distributed computational grids.  ...  Acknowledgements The authors would like to gratefully thank LBNL, LLNL, and SDSC for providing the batch job trace files.  ... 
doi:10.1145/1048935.1050194 dblp:conf/sc/ShanOB03 fatcat:n7nodwsxdvfgtixrozk2vbcq7i

Parallel metaheuristics: recent advances and new trends

Enrique Alba, Gabriel Luque, Sergio Nesmachnow
2012 International Transactions in Operational Research  
In this article, we review these recent research areas in connection to parallel metaheuristics, as well as we identify future trends and possible open research lines for groups and PhD students.  ...  However, porting existing algorithms to new hardware is not enough as a scientific goal, therefore researchers are looking for new parallel optimization and learning models that are targeted to these new  ...  Acknowledgements This work has been partially funded by the Spanish Ministry of Science and Innovation and FEDER under contract TIN2008-06491-C04-01 (the MSTAR project) and TIN2011-28194 (the RoadME project  ... 
doi:10.1111/j.1475-3995.2012.00862.x fatcat:7eqilt2zwbeyhfst5btn6ssvga

Computing resource transformation, consolidation and decomposition in hybrid clouds

Jinho Hwang
2015 2015 11th International Conference on Network and Service Management (CNSM)  
Also, cloud customers want to migrate between cloud providers with different requirements such as cost, performance, and manageability.  ...  With the promise of providing flexible and elastic computing resources on demand, the cloud computing has been attracting enterprises and individuals to migrate workloads in the legacy environment to the  ...  For example, if one is using Linux or similar operating systems, shell is the easiest and the most convenient way to program and run.  ... 
doi:10.1109/cnsm.2015.7367350 dblp:conf/cnsm/Hwang15 fatcat:z7gau6sn7bbbzcef35x5imnjwm

Flexible Management on BSP Process Rescheduling: Offering Migration at Middleware and Application Levels

Lucas Graebin, Rodrigo Da Rosa Righi
2012 Journal of Applied Computing Research  
It was designed to work on grid computing environments and offers an interface that follows the BSP (Bulk Synchronous Parallel) style. jMigBSP's main contribution focuses on the rescheduling facility in  ...  This article describes the rationales for developing jMigBSP -a Java programming library that offers object rescheduling.  ...  Introduction Load balancing is a key issue for getting performance on heterogeneous and dynamic distributed environments (El Kabbany et al., 2011) .  ... 
doi:10.4013/jacr.2011.12.03 fatcat:mvkhoyji6nd6ribept5is6oj3i
« Previous Showing results 1 — 15 out of 26,069 results