26 Hits in 5.4 sec

Trace-based simulations of processor co-allocation policies in multiclusters

A.I.D. Bucur, D.H.J. Epema
High Performance Distributed Computing, 2003. Proceedings. 12th IEEE International Symposium on  
In this paper we study the performance of several scheduling policies for co-allocating unordered requests in multiclusters with a workload derived from the DAS.  ...  of processors to single jobs in multiple clusters, may be required.  ...  Showing the viability of co-allocation in such systems may be regarded as a first step in assessing the benefit of co-allocation in more general grid environments.  ... 
doi:10.1109/hpdc.2003.1210017 dblp:conf/hpdc/BucurE03 fatcat:rqjqvxgnsnev3h5feqmkaj7r3u

The Influence of the Structure and Sizes of Jobs on the Performance of Co-allocation [chapter]

Anca I.D. Bucur, Dick H.J. Epema
2000 Lecture Notes in Computer Science  
These types of job requests are differentiated by the restrictions they impose on the scheduler and by the form of co-allocation used.  ...  We propose a model for scheduling rigid jobs consisting of multiple components in multicluster systems by pure space sharing, based on the Distributed ASCI Supercomputer.  ...  We also intend to do simulations and performance measurements using traces from real multicluster systems instead of theoretical distributions.  ... 
doi:10.1007/3-540-39997-6_10 fatcat:hno27ri4fngfloj6escmm2jshy

The Influence of Communication on the Performance of Co-allocation [chapter]

A.I.D. Bucur, D.H.J. Epema
2001 Lecture Notes in Computer Science  
allocation of processors in different clusters.  ...  We simulate multicluster systems with the FCFS policy-and with two policies for placing a flexible request, one tries to balance cluster loads and one tries to fill clusters completely-to determine the  ...  In this paper, we assess the performance of co-allocation in a model of multicluster systems based on the DAS, with a special emphasis on the effect of the differences in communication speeds between intracluster  ... 
doi:10.1007/3-540-45540-x_5 fatcat:5qqfqrhykva3tmss6r67y2wq74

Scheduling Policies for Processor Coallocation in Multicluster Systems

Anca I.D. Bucur, Dick H.J. Epema
2007 IEEE Transactions on Parallel and Distributed Systems  
In this paper, we propose four scheduling policies for processor coallocation in multiclusters, and we assess with simulations their performance under a wide variety of parameter settings.  ...  Jobs running on multiclusters and, more generally, in grids, may require (processor) coallocation, i.e., the simultaneous allocation of resources (processors) in different clusters or subsystems of a grid  ...  TRACE-BASED SIMULATIONS In this section, we assess the performance of the Global Scheduler, the Local Schedulers (that is, LS-DO), and the Local Priority (LP-GF) policies with simulations based on traces  ... 
doi:10.1109/tpds.2007.1036 fatcat:r5kmuyp2tzac3p74zbxlenbb3u

Scheduling of scientific workflow in non-dedicated heterogeneous multicluster platform

Jinghui Zhang, Junzhou Luo, Fang Dong
2013 Journal of Systems and Software  
To address this problem, we study the scheduling of scientific workflow in a non-dedicated heterogeneous multicluster platform aimed at minimizing the makespan for workflow execution.  ...  Therefore, scheduling such scientific workflow in a multicluster platform becomes a challenging task.  ...  Acknowledgements This work is supported by National Key Basic Research Program of China under Grants No. 2010CB328104, National Natural Science Foundation of China under Grants No. 61070161, No. 61003257  ... 
doi:10.1016/j.jss.2012.10.029 fatcat:vdmyaesghjgfneecdk26rxq4ia

Efficient Methods for Scheduling Jobs in a Simulation Model Using a Multicore Multicluster Architecture

Francisca A. P. Pinto, Henrique J. A. Holanda, Carla K. de M. Marques, Giovanni C. Barroso
2019 Computing and informatics  
Over the past decade, the fast advance of network technologies, hardware and middleware, as well as software resource sophistication has contributed to the emergence of new computational models.  ...  Consequently, there was a capacity increasing for efficient and effective use of resources distributed aiming to  ...  Acknowledgment We thank the CAPES (Coordination for the Improvement of Higher Education Personnel) for the financial support, the National Postdoctoral Program Department of Computer Science -UERN/UFERSA  ... 
doi:10.31577/cai_2019_2_390 fatcat:xbunvcfmcbf77ota4sqcjeodjy

Ensuring fairness among participating clusters during multi-site parallel job scheduling

W.M. Jones, W.B. Ligon
2006 12th International Conference on Parallel and Distributed Systems - (ICPADS'06)  
In this paper, we demonstrate that these multi-site scheduling techniques can be successfully integrated with fairness policies to ensure that participation in the multicluster is beneficial under extremely  ...  Multi-cluster schedulers can dramatically improve average job turn-around time performance by making use of fragmented node resources available throughout the grid.  ...  They therefore form a basis for the best-case execution time of a given job when it is co-allocated in the multicluster. In particular, T E = T P +T C .  ... 
doi:10.1109/icpads.2006.44 dblp:conf/icpads/JonesL06 fatcat:7f7jpuik2jdmtntfht7765kgw4

The impact of information availability and workload characteristics on the performance of job co-allocation in multi-clusters

W.M. Jones, W.B. Ligon, N. Shrivastava
2006 12th International Conference on Parallel and Distributed Systems - (ICPADS'06)  
In this paper, we utilize a bandwidth-centric job communication model that captures the interaction and impact of simultaneously co-allocating jobs across multiple clusters.  ...  In this research, we focus on scheduling strategies that make use of available information such as network link utilization, per-processor bandwidths, and job communication topology in order to make intelligent  ...  They therefore form a basis for the best-case execution time of a given job when it is co-allocated in the multicluster.  ... 
doi:10.1109/icpads.2006.105 dblp:conf/icpads/JonesLS06 fatcat:bp55kulxtfg63ncxptq4b5ouvm

Characterization of Bandwidth-Aware Meta-Schedulers for Co-Allocating Jobs Across Multiple Clusters

William M. Jones, Walter B. Ligon, Louis W. Pang, Dan Stanzione
2005 Journal of Supercomputing  
We explore the interaction of simultaneously co-allocated jobs and the contention they often create in the network infrastructure of a dedicated computational multicluster.  ...  In this paper, we present a bandwidth-centric job communication model that captures the interaction and impact of simultaneously co-allocating jobs across multiple clusters.  ...  Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation.  ... 
doi:10.1007/s11227-005-2337-x fatcat:i2hgjvjw7za7xp4t4sv5wjvzxm

Multisite co-allocation algorithms for computational grid

Weizhe Zhang, A.M.K. Cheng, Mingzeng Hu
2006 Proceedings 20th IEEE International Parallel & Distributed Processing Symposium  
Efficient multisite job scheduling facilitates the cooperation of multi-domain massively parallel processor systems in a computing grid environment.  ...  However, co-allocation, heterogeneity, adaptability, and scalability emerge as tough challenges for the design of multisite job scheduling models and algorithms.  ...  Furthermore, they use the measure-based trace to evaluate the performance of the scheduling policies, design a dynamic co-allocation service and implement multiple components for a realworld wide-area  ... 
doi:10.1109/ipdps.2006.1639652 dblp:conf/ipps/ZhangCH06 fatcat:rl6mnayvxve4tayhmkj3hfmtea

Better Safe than Sorry

Bogdan Ghit, Dick Epema
2017 Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing - HPDC '17  
Then we perform simulations of complete workloads, mimicking the size and operation of a Google cluster, and show that panda provides significant improvements in the average job runtime for wide ranges  ...  These properties form the basis of three checkpointing policies which we incorporate into panda.  ...  To remove the burden of performing sensitivity analysis for each policy, we propose two simple rules of thumb based on the history of job executions in production clusters and in our das multicluster system  ... 
doi:10.1145/3078597.3078600 dblp:conf/hpdc/GhitE17 fatcat:n7xyk65pprg6xd7omtw2fvkttq

A Periodic Portfolio Scheduler for Scientific Computing in the Data Center [chapter]

Kefeng Deng, Ruben Verboon, Kaijun Ren, Alexandru Iosup
2014 Lecture Notes in Computer Science  
Through simulation based on real and synthetic workload traces, we show evidence that portfolio scheduling can automatically select the scheduling policy to match both user and data center objectives,  ...  We design a periodic portfolio scheduler for the workload of the entire data center, and equip it with a portfolio of resource provisioning and allocation policies.  ...  We have evaluated the behavior of our portfolio scheduler through simulations, based on real and synthetic workload traces.  ... 
doi:10.1007/978-3-662-43779-7_9 fatcat:qretxsvayfes3h52p7tjznzeue

Usage SLA-based scheduling in Grids

Catalin L. Dumitrescu, Ioan Raicu, Ian Foster
2007 Concurrency and Computation  
We also address, in great detail, the performance of different uSLA-based site selection strategies and the overall performance in scheduling workloads over Grid3 with workload sizes ranging from 10 to  ...  Running workloads in such environments is often a similarly challenging problem owing to the scale of the environment, and to the resource partitioning based on various sharing strategies.  ...  ACKNOWLEDGEMENTS Some of the results in this paper were supported by the NSF Information Technology Research GriPhyN project, under contract ITR-0086044.  ... 
doi:10.1002/cpe.1091 fatcat:6vwckrplufbgjmvqpq6nlm2goy

Failure-aware resource provisioning for hybrid Cloud infrastructure

Bahman Javadi, Jemal Abawajy, Rajkumar Buyya
2012 Journal of Parallel and Distributed Computing  
Simulation results reveal that in a realistic working condition while adopting user estimates for the requests in the provisioning policies, we are able to improve the users' QoS about 32% in terms of  ...  In this paper, we propose a scalable hybrid Cloud infrastructure as well as resource provisioning policies to assure QoS targets of the users.  ...  Based on the results of trace-based simulations, we observe a considerable improvement in terms of violation rate mainly in the case of tight deadlines.  ... 
doi:10.1016/j.jpdc.2012.06.012 fatcat:gskvfdjs6zdz5aopkyeiio5lou

Compiler-Assisted Data Distribution and Network Configuration for Chip Multiprocessors

Yong Li, A. Abousamra, R. Melhem, A. K. Jones
2012 IEEE Transactions on Parallel and Distributed Systems  
At run time, symbolic MMAPs are resolved and used by a partitioning algorithm to choose a partition of allocated memory blocks among the forked threads in the analyzed application.  ...  Based on the partition, the communication pattern of the application can be extracted. We demonstrate how this information can be used in an experimental architecture to accelerate applications.  ...  Another technique of Chu and Malhke [9] describes an approach to partition data objects among multicluster processors.  ... 
doi:10.1109/tpds.2011.279 fatcat:nhq7im22xjazddsfmgrwdlxofe
« Previous Showing results 1 — 15 out of 26 results