The Effects of Heterogeneity on Asynchronous Panmictic Genetic Search

Abstract

Abstract. Research scientists increasingly turn to large-scale heterogeneous environments such as computational grids and the Internet based facilities to satisfy their rapidly growing computational needs. The increasing complexity of the scientific models and rapid collection of new data are drastically outpacing the advances in processor speed while the cost of supercomputing environments remains relatively high. However, the heterogeneity and unreliability of these environments, especially the Internet, make scalable and fault tolerant search methods indispensable to effective scientific model verification. An effective search method for these types of environments is asynchronous genetic search, where a population continuously evolves based on asynchronously generated and received results. However, it is unclear what effect heterogeneity has on this type of search. For example, results received from slower workers may turn out to be obsolete or less beneficial than results calculated by faster workers. This paper examines the effect of heterogeneity on asynchronous panmictic (single population) genetic search for two different scientific applications, one used by astronomers to model the Milky Way galaxy and another by particle physicists to determine the existence of theory predicted, yet unobserved particles such as missing baryons. Results show that for both applications results received from slower workers while overall less beneficial are still useful. Additionally, a modification of asynchronous genetic search shows that different parameter generation strategies change their effectiveness over the course of the search 1 .

Optimizing long short-term memory recurrent neural networks using ant colony optimization to predict turbine engine vibration

“…The distributed algorithm utilizes an asynchronous master worker approach, which has been shown to provide performance and scalability over iterative approaches in evolutionary algorithms [58,59]. This approach provides an additional benefit in that it is automatically load balanced -workers request and receive new LSTM RNNs when they have completed training previous ones, without blocking on results from other workers.…”

Evolutionary Computation in Combinatorial Optimization

“…It is implemented in C++ and MPI, allowing easy use on clusters and supercomputers, and also provides support for systems with multiple graphical processing units. Further, TAO has shown that performing evolutionary algorithms asynchronously can provide significant improvements to performance and scalability over iterative approaches [13,39]. TAO is open source and freely available on GitHub, allowing easy use and extensibility 1 , and the presented ACO strategy has been included in that repository.…”

Evolving N-Body Simulations to Determine the Origin and Structure of the Milky Way Galaxy's Halo Using Volunteer Computing

“…The disruption depends on initial properties of the dwarf galaxy and on the gravitational potential of the Milky Way, which is primarily due to dark matter. Asynchronous evolutionary algorithms (AEAs) have been used successfully to find the optimal input parameters for the probabilistic sampling method [3], [4], [3], [5], [6], and this work expands these AEAs to find the optimal input parameters for the N-body simulations, which will in turn provide new understanding of the origin and structure of the Milky Way.…”