Cloud Driven Design of a Distributed Genetic Programming Platform

Derby, Owen; Veeramachaneni, Kalyan; O’Reilly, Una-May

doi:10.1007/978-3-642-37192-9_51

Cited by 6 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…XMLHttpRequest is an API that allows JavaScript code to send HTTP and HTTPS requests in the background to the same domain name as the one from which a web page was downloaded from 7 . This protocol has many advantages: it is mature, it is supported by every modern browser, and it is simple to use.…”

Section: Xmlhttprequestmentioning

confidence: 99%

“…Moreover, WebSockets are not interoperable with classic TCP sockets, and currently they are not implemented in all web browsers. 7 There are also ways to reach other domains name using CORS (http://www.w3.org/TR/cors/) policies. 8 http://www.w3.org/TR/websockets/ Socket.io Finally, Socket.io 9 , is a bleeding edge technology which solves the few shortcomings of WebSockets, while also providing compatibility with a diverse set of programming languages.…”

Section: Websocketsmentioning

confidence: 99%

“…Specifically, we have included virtual machines provided by Amazon EC2 (as also done, for instance, in [11]), but other virtual machine providers, such as Microsoft Azure 15 or the Google cloud platform 16 would also be straightforward to incorporate. Another possibility would be to set up a private cloud system using OpenStack, as done in [7,29].…”

Section: Leveraging the "Cloud"mentioning

confidence: 99%

“…Distributed GP was also contributed, more recently, in [7,29]: the proposed P2P framework, called FlexGP, is based on the OpenStack 2 open-source cloud computing platform and makes use of a simple island-based system where all the peer nodes run independent copies of the GP-based regression algorithm, each one using different parameters and learning from different subsets of the original dataset. The best models obtained in each node are eventually fused offline to provide the user with an optimal meta-model for the regression task at hand (the system is, therefore, particularly suitable for solving large regression problems).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Seamless Peer and Cloud Evolution Framework

Leclerc

Auerbach

Iacca

et al. 2016

Proceedings of the Genetic and Evolutionary Computation Conference 2016

View full text Add to dashboard Cite

Evolutionary algorithms are increasingly being applied to problems that are too computationally expensive to run on a single personal computer due to costly fitness function evaluations and/or large numbers of fitness evaluations. Here, we introduce the Seamless Peer And Cloud Evolution (SPACE) framework, which leverages bleeding edge web technologies to allow the computational resources necessary for running large scale evolutionary experiments to be made available to amateur and professional researchers alike, in a scalable and cost-effective manner, directly from their web browsers. The SPACE framework accomplishes this by distributing fitness evaluations across a heterogeneous pool of cloud compute nodes and peer computers. As a proof of concept, this framework has been attached to the RoboGen TM open-source platform for the co-evolution of robot bodies and brains, but importantly the framework has been built in a modular fashion such that it can be easily coupled with other evolutionary computation systems.

show abstract

Section: Xmlhttprequestmentioning

confidence: 99%

Section: Websocketsmentioning

confidence: 99%

Section: Leveraging the "Cloud"mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Seamless Peer and Cloud Evolution Framework

Leclerc

Auerbach

Iacca

et al. 2016

Proceedings of the Genetic and Evolutionary Computation Conference 2016

View full text Add to dashboard Cite

show abstract

“…Thus, new PEAs have been designed to take advantage of cloud computing systems [Arenas et al(2011), Derby et al(2013)], grid systems [Lim et al(2007)], or volunteer computing systems [Desell et al(2010), Biazzini and Montresor(2013), O'Reilly et al(2013)]. The latter has become popular since volunteer systems provide a unique infrastructure of massively large and potentially free resources for tackling challenging problems; the Milky-Way@HOME project [Desell et al(2008)], for example, is a time-consuming volunteer-based application that tries to find an optimal fit of three-dimensional models of the Milky Way galaxy to star data observed.…”

Section: Related Workmentioning

confidence: 99%

Designing robust volunteer-based evolutionary algorithms

Laredo

Bouvry

González³

et al. 2014

Genet Program Evolvable Mach

View full text Add to dashboard Cite

This paper tackles the design of scalable and fault-tolerant evolutionary algorithms computed on volunteer platforms. These platforms aggregate computational resources from contributors all around the world. Given that resources may join the system only for a limited period of time, the challenge of a volunteer-based evolutionary algorithm is to take advantage of a large amount of computational power that in turn is volatile. The paper analyzes first the speed of convergence of massively parallel evolutionary algorithms. Then, it provides some guidance about how to design efficient policies to overcome the algorithmic loss of quality when the system undergoes high rates of transient failures, i.e. computers fail only for a limited period of time and then become available again. In order to provide empirical evidence, experiments were conducted for two well-known problems which require large population sizes to be solved, the first based on a genetic algorithm and the second on genetic programming. Results show that, in general, evolutionary algorithms undergo a graceful degradation under the stress of losing computing nodes. Additionally, new available nodes can also contribute to improving the search process. Despite losing up to 90% of the initial computing resources, volunteer-based evolutionary algorithms can find the same solutions in a failure-prone as in a failure-free run.

show abstract