Flexible self-adjustment of task deployment in dynamic environments

This paper presents an investigation into dynamic self-adjustment of task deployment and other aspects of self-management, through the embedding of multiple policies. Non-dedicated loosely-coupled computing environments, such as clusters and grids are increasingly popular platforms for parallel processing. These abundant systems are highly dynamic environments in which many sources of variability affect the run-time efficiency of tasks. The dynamism is exacerbated by the incorporation of mobile devices and wireless communication. This paper proposes an adaptive strategy for the flexible run-time deployment of tasks; to continuously maintain efficiency despite the environmental variability. The strategy centres on policy-based scheduling which is informed by contextual and environmental inputs such as variance in the round-trip communication time between a client and its workers and the effective processing performance of each worker. A self-management framework has been implemented for evaluation purposes. The framework integrates several policy-controlled, adaptive services with the application code, enabling the run-time behaviour to be adapted to contextual and environmental conditions. Using this framework, an exemplar self-managing parallel application is implemented and used to investigate the extent of the benefits of the strategy

Policy-based techniques for self-managing parallel applications

This paper presents an empirical investigation of policy-based self-management techniques for parallel applications executing in loosely-coupled environments. The dynamic and heterogeneous nature of these environments is discussed and the special considerations for parallel applications are identified. An adaptive strategy for the run-time deployment of tasks of parallel applications is presented. The strategy is based on embedding numerous policies which are informed by contextual and environmental inputs. The policies govern various aspects of behaviour, enhancing flexibility so that the goals of efficiency and performance are achieved despite high levels of environmental variability. A prototype self-managing parallel application is used as a vehicle to explore the feasibility and benefits of the strategy. In particular, several aspects of stability are investigated. The implementation and behaviour of three policies are discussed and sample results examined.

Load-balancing for parallel adaptive unstructured grids

A parallel method for the dynamic partitioning of unstructured meshes is outlined. The method includes diffusive load-balancing techniques and an iterative optimisation technique known as relative gain optimisationwhich both balances theworkload and attempts to minimise the interprocessor communications overhead. It can also optionally include amultilevel strategy. Experiments on a series of adaptively refined meshes indicate that the algorithmprovides partitions of an equivalent or higher quality to static partitioners (which do not reuse the existing partition) and much more rapidly. Perhaps more importantly, the algorithm results in only a small fraction of the amount of data migration compared to the static partitioners.

Parallel dynamic load-balancing for adaptive unstructured meshes

This chapter describes a parallel optimization technique that incorporates a distributed load-balancing algorithm and provides an extremely fast solution to the problem of load-balancing adaptive unstructured meshes. Moreover, a parallel graph contraction technique can be employed to enhance the partition quality and the resulting strategy outperforms or matches results from existing state-of-the-art static mesh partitioning algorithms. The strategy can also be applied to static partitioning problems. Dynamic procedures have been found to be much faster than static techniques, to provide partitions of similar or higher quality and, in comparison, involve the migration of a fraction of the data. The method employs a new iterative optimization technique that balances the workload and attempts to minimize the interprocessor communications overhead. Experiments on a series of adaptively refined meshes indicate that the algorithm provides partitions of an equivalent or higher quality to static partitioners (which do not reuse the existing partition) and much more quickly. The dynamic evolution of load has three major influences on possible partitioning techniques; cost, reuse, and parallelism. The unstructured mesh may be modified every few time-steps and so the load-balancing must have a low cost relative to that of the solution algorithm in between remeshing.

Load balancing for aspect ratio in adaptive finite element simulations

In parallel adaptive finite element simulations the work load on the individual processors may change frequently. To (re)distribute the load evenly over the processors a load balancing heuristic is needed. Common strategies try to minimise subdomain dependencies by optimising the cutsize of the partitioning. However for certain solvers cutsize only plays a minor role, and their convergence is highly dependent on the subdomain shapes. Degenerated subdomain shapes cause them to need significantly more iterations to converge. In this work a new parallel load balancing strategy is introduced which directly addresses the problem of generating and conserving reasonably good subdomain shapes in a dynamically changing Finite Element Simulation. Geometric data is used to formulate several cost functions to rate elements in terms of their suitability to be migrated. The well known diffusive method which calculates the necessary load flow is enhanced by weighting the subdomain edges with the help of these cost functions. The proposed methods have been tested and results are presented.