Exploiting Load Imbalance Patterns for Heterogeneous Cloud Computing Platforms

Cloud computing providers offer a variety of instance sizes, types, and configurations that have different prices but can interoperate. As many parallel applications have heterogeneous computational demands, these different instance types can be exploited to reduce the cost of executing a parallel application while maintaining an acceptable performance. In this paper, we perform an analysis of load imbalance patterns with an intentionally-imbalanced artificial benchmark to discover which patterns can benefit from a heterogeneous cloud system. Experiments with this artificial benchmark as well as applications from the NAS Parallel Benchmark suite show that the price of executing an imbalanced application can be reduced substantially on a heterogeneous cloud for a variety of imbalance patterns, while maintaining acceptable performance. By using a heterogeneous cloud, cost efficiency was improved by up to 63%, while performance was reduced by less

[1]  Jian Li,et al.  Cost-efficient task scheduling for executing large programs in the cloud , 2013, Parallel Comput..

[2]  Philippe Olivier Alexandre Navaux,et al.  Exploiting Price and Performance Tradeoffs in Heterogeneous Clouds , 2017, UCC.

[3]  Stephen P. Crago,et al.  Heterogeneous Cloud Computing: The Way Forward , 2015, Computer.

[4]  Philippe Olivier Alexandre Navaux,et al.  Locality and Balance for Communication-Aware Thread Mapping in Multicore Systems , 2015, Euro-Par.

[5]  Katherine Yelick,et al.  Hierarchical Work Stealing on Manycore Clusters , 2011 .

[6]  John P. Morrison,et al.  Managing and Unifying Heterogeneous Resources in Cloud Environments , 2017, CLOSER.

[7]  Hsien-Hsin S. Lee,et al.  Using Mathematical Modeling in Provisioning a Heterogeneous Cloud Computing Environment , 2011, Computer.

[8]  Philippe Olivier Alexandre Navaux,et al.  High Performance Computing in the cloud: Deployment, performance and cost efficiency , 2012, 4th IEEE International Conference on Cloud Computing Technology and Science Proceedings.

[9]  Nancy M. Amato,et al.  Quantifying the effectiveness of load balance algorithms , 2012, ICS '12.

[10]  Kevin Skadron,et al.  Rodinia: A benchmark suite for heterogeneous computing , 2009, 2009 IEEE International Symposium on Workload Characterization (IISWC).

[11]  Asser N. Tantawi,et al.  See Spot Run: Using Spot Instances for MapReduce Workflows , 2010, HotCloud.

[12]  Manuel Prieto,et al.  Survey of scheduling techniques for addressing shared resources in multicore processors , 2012, CSUR.

[13]  Jack J. Dongarra,et al.  The LINPACK Benchmark: past, present and future , 2003, Concurr. Comput. Pract. Exp..

[14]  Boon Thau Loo,et al.  Exploiting cloud heterogeneity for optimized cost/performance MapReduce processing , 2014, CloudDP '14.

[15]  Philippe Olivier Alexandre Navaux,et al.  Automatic Communication Optimization of Parallel Applications in Public Clouds , 2016, 2016 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid).

[16]  Philippe Olivier Alexandre Navaux,et al.  Leveraging Cloud Heterogeneity for Cost-Efficient Execution of Parallel Applications , 2017, Euro-Par.

[17]  David H. Bailey,et al.  The Nas Parallel Benchmarks , 1991, Int. J. High Perform. Comput. Appl..

[18]  Boon Thau Loo,et al.  Exploiting Cloud Heterogeneity to Optimize Performance and Cost of MapReduce Processing , 2015, PERV.

[19]  Changjun Jiang,et al.  Improving Performance of Heterogeneous MapReduce Clusters with Adaptive Task Tuning , 2017, IEEE Transactions on Parallel and Distributed Systems.

[20]  Vivien Quéma,et al.  Traffic management: a holistic approach to memory placement on NUMA systems , 2013, ASPLOS '13.

[21]  Philippe Olivier Alexandre Navaux,et al.  Locality vs. Balance: Exploring Data Mapping Policies on NUMA Systems , 2015, 2015 23rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing.