9,985 Hits in 8.0 sec

Optimal Load Balancing and Assessment of Existing Load Balancing Criteria [article]

Anthony Boulmier, Nabil Abdennadher, Bastien Chopard
2021 arXiv   pre-print
Hence, load balancing techniques are used to distribute the workload evenly to maximize performance. A key challenge is to know when to use load balancing techniques.  ...  This algorithm finds the optimal load balancing scenario in quadratic time while, to the best of our knowledge, this has never been addressed in less than an exponential time.  ...  The irregular workload benchmarks target applications with varying workload per time-step, and where the load imbalance comes from both the problem in itself and the parallelization. our criterion, and  ... 
arXiv:2104.01688v1 fatcat:jp4srrwfwngihlbjk5rf7v6bbu

Passive/Active Load Balancing with Informed Node Placement in DHTs [chapter]

Mikael Högqvist, Nico Kruber
2009 Lecture Notes in Computer Science  
Support for range queries introduces new challenges to load balancing since both the key and workload distribution can be non-uniform.  ...  We build on previous work based on the power of choice to present algorithms suitable for active and passive load balancing that adapt to both the key and workload distribution.  ...  This work is partially funded by the European Commission through the SELFMAN project with contract number 034084.  ... 
doi:10.1007/978-3-642-10865-5_9 fatcat:faznsm45e5fthe62iv44ohspfu

Efficient Workload Balancing on Heterogeneous GPUs using MixedInteger Non-Linear Programming

Chih-Sheng Lin, Chih-Wei Hsieh, Hsi-Ya Chang, Pao-Ann Hsiung
2014 Journal of Applied Research and Technology  
However, for application designers, computational workload still needs to be distributed to heterogeneous GPUs manually and remains inefficient.  ...  Moreover, our method only requires a few overhead while achieving high performance and load balancing.  ...  Acknowledgments We are grateful to the National Center for High-Performance Computing for computer time and facilities.  ... 
doi:10.1016/s1665-6423(14)71676-1 fatcat:dfp72j3e2vcbpejckd2r7wibby

On the Benefits of Anticipating Load Imbalance for Performance Optimization of Parallel Applications

Anthony Boulmier, Franck Raynaud, Nabil Abdennadher, Bastien Chopard
2019 2019 IEEE International Conference on Cluster Computing (CLUSTER)  
So far, the standard load balancing method consists in distributing the workload evenly between PEs and, when load imbalance appears, redistributing the extra load from overloaded PEs to underloaded PEs  ...  In parallel iterative applications, computational efficiency is essential for addressing large problems. Load imbalance is one of the major performance degradation factors of parallel applications.  ...  Most of the time, load imbalance arises from problem, algorithmic, and systemic characteristics [1] . To mitigate load imbalance, load balancing (LB) techniques have been developed over the years.  ... 
doi:10.1109/cluster.2019.8890998 dblp:conf/cluster/BoulmierRAC19 fatcat:azbljgzlube4hpe5ojs2ngophm

An efficient load balancing LQR controller in parallel database queries under random perturbations

Anastasios Gounaris, Christos A. Yfoulis, Norman W. Paton
2009 2009 IEEE International Conference on Control Applications  
This work investigates the problem of dynamic, intra-query load balancing in parallel database queries across heterogeneous nodes in a way that takes into account the inherent cost of adaptations and thus  ...  The latter may lead to serious performance degradation in several cases, such as periodic and random imbalances.  ...  performance under time-varying imbalance conditions.  ... 
doi:10.1109/cca.2009.5280982 dblp:conf/IEEEcca/GounarisYP09 fatcat:u6gq3pm6yjehrizdz7hj3ibeum

Toward Dynamic Load Balancing across OpenMP Thread Teams for Irregular Workloads

Xiong Xiao, Shoichi Hirasawa, Hiroyuki Takizawa, Hiroaki Kobayashi
2017 International Journal of Networking and Computing  
load balancing across teams, with considering scheduling overhead.  ...  The results show that such a dynamic load balancing mechanism can improve the 387 Toward Dynamic Load Balancing across OpenMP Thread Teams for Irregular Workloads performance by up to 14% compared to static  ...  for Scientific Research(B) 16H02822.  ... 
doi:10.15803/ijnc.7.2_387 fatcat:foyiqnt3ubgvdjenkbck6hzmf4

A Comparison of Implementation Strategies for Nonuniform Data-Parallel Computations

Salvatore Orlando, Raffaele Perego
1998 Journal of Parallel and Distributed Computing  
Workload imbalance can be mitigated by cyclically distributing data and associated computations, or by employing adaptive strategies which build a more balanced schedule at run{time, on the basis of the  ...  W e fed the benchmark code with synthetic input data sets built on the basis of a load imbalance model and we report and compare the results obtained.  ...  applications), or iterated loops where the workload distribution is highly varying.  ... 
doi:10.1006/jpdc.1998.1456 fatcat:dpu3rbatdned7evd5pvvmmnsii

Parallel Stream Processing Against Workload Skewness and Variance [article]

Junhua Fang, Rong Zhang, Tom Z.J.Fu, Zhenjie Zhang, Aoying Zhou, Junhua Zhu
2016 arXiv   pre-print
While randomized hashing on the keys is capable of balancing the workload for key-based partitioning when the keys generally follow a static distribution, it is likely to generate poor balancing performance  ...  We formulate the rebalance operation as an optimization problem, with multiple objectives on minimizing state migration costs, controlling the size of the routing table and breaking workload imbalance  ...  The results of load imbalance in Fig. 7 are presented as the cumulative distribution of average workload among the task instances over 50 time intervals.  ... 
arXiv:1610.05121v2 fatcat:6sz7iormmngaxd7tu5gymclspq

Efficient execution of Time Warp programs on heterogeneous, NOW platforms

C.D. Carothers, R.M. Fujimoto
2000 IEEE Transactions on Parallel and Distributed Systems  
PCS: cluster allocation over time with combination of four time-varying (150/50 second case), external tasks on PE 1 and static internal workload on PE 0. For GTW with BGE, Â H:HS, and T schedule S.  ...  Index TermsÐDiscrete event simulation, distributed simulation, network of workstations (NOW), time warp, dynamic load balancing. Fig. 10.  ...  Additionally, Avril and Tropper [2] present a scheme for dynamically load balancing Time Warp programs with irregular, internal workloads.  ... 
doi:10.1109/71.841745 fatcat:ctcpguvr3rca3g5m6kni6jhbye

Qualitative Parametric Comparison Of Load Balancing Algorithms In Parallel And Distributed Computing Environment

Amit Chhabra, Gurvinder Singh, Sandeep Singh Waraich, Bhavneet Sidhu, Gaurav Kumar
2008 Zenodo  
time, Dynamic load balancing algorithms (DLB) are adaptive to changing situations and take decisions at run time.  ...  Such imbalances in system load suggest that performance can be improved by either transferring jobs from the currently heavily loaded hosts to the lightly loaded ones or distributing load evenly/fairly  ...  System administrators have attempted static balancing of workloads with user accounts for a long time.  ... 
doi:10.5281/zenodo.1059751 fatcat:nswd2a2q3bgcjgzweptk6obynm

SUPPLE: An efficient run-time support for non-uniform parallel loops

Salvatore Orlando, Raffaele Perego
1999 Journal of systems architecture  
If the original load distribution is not uniform, the data layout is thus modi ed in order to balance the processor loads.  ...  SUPPLE always tries to overlap communications with useful computations by reordering loop iterations and prefetching remote ones in the case of workload imbalance.  ...  However, static optimizations like these may b e unsuccessful when the workload distribution varies unpredictably at run time.  ... 
doi:10.1016/s1383-7621(98)00071-x fatcat:kiu4ma2g5bf2na72mnmf6xsatq

Scalable and robust key group size estimation for reducer load balancing in MapReduce

Wei Yan, Yuan Xue, Bradley Malin
2013 2013 IEEE International Conference on Big Data  
We introduce a sketch-based data structure for capturing MapReduce key group size statistics and present an optimal packing algorithm which assigns the key groups to the reducers in a load balancing manner  ...  However, in the real world, data is often highly skewed, which may cause workload imbalance among parallel running tasks.  ...  To cope with load imbalance, these approaches first profile the input data to collect distribution information and then utilize these information to achieve reduce-phase load balancing.  ... 
doi:10.1109/bigdata.2013.6691568 dblp:conf/bigdataconf/YanXM13 fatcat:dwgjeqfeyjgihi36cak54hl2ie

Page 158 of American Society of Civil Engineers. Collected Journals Vol. 8, Issue 3 [page]

1995 American Society of Civil Engineers. Collected Journals  
Variable processor load in a multiuser multitasking en- vironment: Workloads on machines vary with time.  ...  A severe load imbalance will arise when some of the conditions just enumerated are violated. In a distributed GA for structural optimization, the fol- lowing factors lead to a load imbalance: 1.  ... 

Reinforcement Learning for Load-balanced Parallel Particle Tracing [article]

Jiayi Xu, Hanqi Guo, Han-Wei Shen, Mukund Raj, Skylar W. Wurster, Tom Peterka
2022 arXiv   pre-print
Our algorithm improves parallel particle tracing performance in terms of parallel efficiency, load balance, and costs of I/O and communication for evaluations with up to 16,384 processors.  ...  We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems.  ...  [20] proposed a matrix-based optimization algorithm to assign blocks to processes for workload balancing. Dynamic load balancing: Workloads of processes are periodically optimized.  ... 
arXiv:2109.05679v2 fatcat:ug2pvmtslffedfa5nfylgw3qn4

Uncovering CPU load balancing policies with harmony

Joseph T. Meehean, Andrea C. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau, Miron Livny
2013 Proceedings of the ACM International Conference on Computing Frontiers - CF '13  
We demonstrate the effectiveness of Harmony by applying it to the analysis of the load-balancing behavior of three Linux schedulers: O(1), CFS, and BFS.  ...  We introduce Harmony, a system for extracting the multiprocessor scheduling policies from commodity operating systems.  ...  The imbalance between the three loaded sources and the single target is varied along the x-axis; that is, [10 + x, 10 + x, 10 + x, 10]. Figure 7 : 7 Time to Resolve Imbalance.  ... 
doi:10.1145/2482767.2482784 dblp:conf/cf/MeeheanAAL13 fatcat:bpcknrs3mfe4behb5mljx4t7my
« Previous Showing results 1 — 15 out of 9,985 results