Filters








166,661 Hits in 7.5 sec

Time-Sharing Parallel Jobs in the Presence of Multiple Resource Requirements [chapter]

Fabrizio Petrini, Wu-chun Feng
2000 Lecture Notes in Computer Science  
overhead and the scheduling overhead to wake up a sleeping process dominate the communication time on most parallel machines [15] .  ...  These actions are based on local events that occur naturally within communicating applications.  ...  This can have a huge impact on the parallelization of existing legacy codes.  ... 
doi:10.1007/3-540-39997-6_8 fatcat:73bbjxqgm5f6rathp245wuqobq

A Novel, Energy-Aware Task Duplication-Based Scheduling Algorithm of Parallel Tasks on Clusters

Aihua Liang, Yu Pang
2016 Mathematical and Computational Applications  
It has advantages over other algorithms on computation-intensive applications.  ...  To balance the energy and performance, we proposed a novel, energy-aware duplication-based scheduling (NEADS).  ...  Impact of Applications According to the ratio of computation and communication, the features of parallel applications can be classified into computation intensive, communication intensive and synthetic  ... 
doi:10.3390/mca22010002 fatcat:7p5utd4cvbh3bnno6stbuau56u

An energy-based implicit co-scheduling model for Beowulf cluster

Sriprayoonsakul, Uthayopas
2003 Proceedings IEEE International Conference on Cluster Computing CLUSTR-03  
The model proposed falls into a framework of implicit co-scheduling techniques where the scheduler uses the recent traces of communication patterns to periodically calculate the priorities of processes  ...  This paper introduces an implicit co-scheduling model called an energy-based model where a parallel system is considered as an RC electric circuit whose components are processes.  ...  Note that since Predictive co-scheduling uses only the communication vol-Multiple parallel applications running concurrently with multiple parallel programs ume of the process, it cannot improve the performance  ... 
doi:10.1109/clustr.2003.1253336 dblp:conf/cluster/SriprayoonsakulU03 fatcat:evdl5vv6ifgdfcup7rh3phhhdq

Towards an Effective Unified Programming Model for Many-Cores

Ana Lucia Varbanescu, Pieter Hijma, Rob van Nieuwpoort, Henri Bal
2011 2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum  
On the one hand, the increasing variety of platforms and their specific programming models force users to take a hardware-centric approach not only for implementing parallel applications, but also for  ...  On the other hand, to effectively cope with the increased number of large-scale workloads that require parallelization, a portable, application-centric programming model is desirable.  ...  Communication and synchronization is explicit by means of messages and the user has no control over how processes are scheduled. However, it allows other optimizations at a later stage.  ... 
doi:10.1109/ipdps.2011.210 dblp:conf/ipps/VarbanescuHNB11 fatcat:6yp3fhaogfebjgcv2ncryieeme

Optimal configuration of compute nodes for synthetic aperture radar processing [chapter]

Jeffrey T. Muehring, John K. Antonio
1998 Lecture Notes in Computer Science  
In general, the mapping of data and the scheduling of communications impacts network contention of parallel architectures.  ...  The primary goals of many applications implemented on parallel architectures are to reduce latency and minimize interprocessor communication time (IPC) while maximizing throughput.  ...  Additionally, Java can be used to create both applications (i.e., a program that executes on a local computer) and applets (i.e., an application that is designed to be transmitted over the Internet and  ... 
doi:10.1007/3-540-64359-1_765 fatcat:l63xljjssfddfc6luxqgohfrqy

EAD and PEBD: Two Energy-Aware Duplication Scheduling Algorithms for Parallel Tasks on Homogeneous Clusters

Ziliang Zong, Adam Manzanares, Xiaojun Ruan, Xiao Qin
2011 IEEE transactions on computers  
On one hand, high performance is certainly an important consideration in designing clusters to run parallel applications.  ...  To achieve the goal of optimizing both performance and energy efficiency in clusters, in this paper, we propose two energy-efficient duplication-based scheduling algorithms-Energy-Aware Duplication (EAD  ...  Their valuable discussions and thoughts have tremendously helped in improving the quality of this paper. The work reported in this paper was supported by the US National Science  ... 
doi:10.1109/tc.2010.216 fatcat:w23gzty4gngpfdnp3hiospywr4

Performance evaluation of scheduling applications with DAG topologies on multiclusters with independent local schedulers

Ligang He, S.A. Jarvis, D.P. Spooner, G.R. Nudd
2006 Proceedings 20th IEEE International Parallel & Distributed Processing Symposium  
The results show that compared with conventional DAG mapping policies, the new scheme that we present can significantly improve the scheduling performance of a DAG-based application in terms of the schedule  ...  After mapping, the tasks (in the DAG-based application) to be executed at each computational resource are determined.  ...  The impact of the ratio of task volume to message volume on the schedule length A further observation is that the advantage of MDM over the base line becomes less prominent as the ratio of task volume  ... 
doi:10.1109/ipdps.2006.1639635 dblp:conf/ipps/HeJSN06 fatcat:dsqo5lhtazgsnnt6j7xcuu34na

Improving the Scalability of Parallel Jobs by adding Parallel Awareness to the Operating System

Terry Jones, Paul Tomlinson, Mark Roberts, Shawn Dawson, Rob Neely, William Tuel, Larry Brenner, Jeffrey Fier, Robert Blackmore, Patrick Caffrey, Brian Maskell
2003 Proceedings of the 2003 ACM/IEEE conference on Supercomputing - SC '03  
A parallel application benefits from scheduling policies that include a global perspective of the application's process working set.  ...  Our results indicate a speedup of over 300% on synchronizing collectives.  ...  Figure 2 : 2 Bulk-Synchronous SPMD Model of parallel application. Each process of a parallel job executes on a separate processor and alternates between computation and communication phases.  ... 
doi:10.1145/1048935.1050161 dblp:conf/sc/JonesDNTBFBCMTR03 fatcat:utaonbdorveyrjgz4h2rvitcd4

Effective distributed scheduling of parallel workloads

Andrea C. Dusseau, Remzi H. Arpaci, David E. Culler
1996 Proceedings of the 1996 ACM SIGMETRICS international conference on Measurement and modeling of computer systems - SIGMETRICS '96  
Local Scheduling The policies of the local scheduler have a significant impact on the performance of parallel applications.  ...  We are able to explore the impact of conditions such as the characteristics of the parallel applications, the performance of the communication network, and the cost of a contextswitch.  ... 
doi:10.1145/233013.233020 dblp:conf/sigmetrics/Arpaci-DusseauAC96 fatcat:w4xtgqhnibd77be6t5gorf6mse

Effective distributed scheduling of parallel workloads

Andrea C. Dusseau, Remzi H. Arpaci, David E. Culler
1996 Performance Evaluation Review  
Local Scheduling The policies of the local scheduler have a significant impact on the performance of parallel applications.  ...  We are able to explore the impact of conditions such as the characteristics of the parallel applications, the performance of the communication network, and the cost of a contextswitch.  ... 
doi:10.1145/233008.233020 fatcat:4263wilhlrhclhuymxn3msgmj4

Mapping DAG-based applications to multiclusters with background workload

L. He, S.A. Jarvis, D.P. Spooner, D. Bacigalupo, G. Tan, G.R. Nudd
2005 CCGrid 2005. IEEE International Symposium on Cluster Computing and the Grid, 2005.  
The results show that compared with conventional DAG mapping policies, the new scheme that we present can significantly improve the scheduling performance of a DAG-based application in terms of the schedule  ...  After mapping, the tasks (in the DAG-based application) to be executed at each computational resource are determined.  ...  The impact of the ratio of task volume to message volume on the schedule length A further observation is that the advantage of MDM over the base line becomes less prominent as the ratio of task volume  ... 
doi:10.1109/ccgrid.2005.1558651 dblp:conf/ccgrid/HeJSBTN05 fatcat:tqn73eepgrfc5ixbtc5chejjpm

Scheduling Workflows in Multi-cluster Environments

S. L. Stanzani, L. M. Sato, M. A. S. Netto
2013 2013 27th International Conference on Advanced Information Networking and Applications Workshops  
We evaluated MCA-HEFT against the Mixed-parallel Heterogeneous Earliest Finish Time (M-HEFT) heuristic, which is one of the most well-known workflow scheduling heuristics in literature.  ...  Scientific applications modeled as workflows can exhibit both task and data parallelism.  ...  Based on the following parameters the generator defines values for computational cost, communication cost, and parallel fraction of each parallel task of the workflow.  ... 
doi:10.1109/waina.2013.30 dblp:conf/aina/StanzaniSN13 fatcat:wyobt5ck6ffhpgykkreslj6e6q

Use of run time predictions for automatic co-allocation of multi-cluster resources for iterative parallel applications

Marco A.S. Netto, Christian Vecchiola, Michael Kirley, Carlos A. Varela, Rajkumar Buyya
2011 Journal of Parallel and Distributed Computing  
We performed experiments using an iterative parallel application, which consists of benchmark multiobjective problems, with both synchronous and asynchronous communication models on Grid'5000.  ...  These static requests, defined by users, limit the initial scheduling and prevent rescheduling of applications to other resource sets.  ...  We also would like to thank Antonio Lima and Raquel Dias for their assistance with the PEPS, ABC toolbox, and Amber applications.  ... 
doi:10.1016/j.jpdc.2011.05.007 fatcat:mwlh4vk2mfagvnrb7ltxchwhnm

Impact of Over-Decomposition on Coordinated Checkpoint/Rollback Protocol [chapter]

Xavier Besseron, Thierry Gautier
2012 Lecture Notes in Computer Science  
In this paper, we study the impact of decomposing an application in much more parallelism that the physical parallelism on the rollback step of fault tolerant coordinated protocols.  ...  We show that the overhead on normal execution remains low for relevant factor of over-decomposition.  ...  Experiments presented in this paper were carried out using the Grid'5000 experimental testbed, being developed under the INRIA AL-ADDIN development action with support from CNRS, RENATER and several Universities  ... 
doi:10.1007/978-3-642-29740-3_36 fatcat:yjq64u4gx5hqrgnwn3cugb3suy

A Novel Approach in Scheduling Of the Real- Time Tasks In Heterogeneous Multicore Processor with Fuzzy Logic Technique For Micro-grid Power Management

Lavanya Dhanesh, P. Murugesan
2018 International Journal of Power Electronics and Drive Systems  
The aim of this paper is to implement a scheduling algorithm based on fuzzy logic for heterogeneous multicore processor for effective micro-grid application.  ...  The main idea is to use two fuzzy logic based scheduling algorithm, first is to assign priority based on execution time and deadline of the task.  ...  Based on six real packet processing applications chosen from Net Bench and Packet Bench, LATA exhibits an average of 36.5% reduction of latency across all applications without substantially degrading the  ... 
doi:10.11591/ijpeds.v9.i1.pp80-88 fatcat:ucvno7g7k5h6zhbhbbkeqlxmii
« Previous Showing results 1 — 15 out of 166,661 results