Creating a concurrent and stateless version of an evolutionary algorithm implies changes in its algorithmic model. From the performance point of view, the main challenge is to balance computation with communication, b...
详细信息
ISBN:
(数字)9783030166922
ISBN:
(纸本)9783030166915;9783030166922
Creating a concurrent and stateless version of an evolutionary algorithm implies changes in its algorithmic model. From the performance point of view, the main challenge is to balance computation with communication, but from the evolutionary point of view another challenge is to keep diversity high so that the algorithm is not stuck in local minima In a concurrent setting, we will have to find the right balance so that improvements in both facets do not cancel out. In this paper we address such an issue, by exploring the space of parameters of a population based concurrent evolutionary algorithm that yields to find out the best combination for a particular problem.
Concurrent languages such as Perl 6 fully leverage the power of current multi-core and hyper-threaded computer architectures, and they include easy ways of automatically parallelizing code. However, to achieve more co...
详细信息
ISBN:
(纸本)9783030003500;9783030003494
Concurrent languages such as Perl 6 fully leverage the power of current multi-core and hyper-threaded computer architectures, and they include easy ways of automatically parallelizing code. However, to achieve more computational capability by using all threads and cores, algorithms need to be redesigned to be run in a concurrent environment;in particular, the use of a reactive, fully functional patterns need to turn the algorithm into a series of stateless steps, with simple functions that receive all the context and map it to the next stage. In this paper, we are going to analyze different versions of these stateless, reactive architectures applied to evolutionary algorithms, assessing how they interact with the characteristics of the evolutionary algorithm itself and show how they improve the scaling behavior and performance. We will use the Perl 6 language, which is a modern, concurrent language that was released recently and is still under very active development.
This paper focus on congestion control for best-effort packet-switching networks, where congested routers use I bit per packet to communicate with sources. Sources adapt their rates according to the sequence of bits r...
详细信息
This paper focus on congestion control for best-effort packet-switching networks, where congested routers use I bit per packet to communicate with sources. Sources adapt their rates according to the sequence of bits received. Routers do not keep per-flow information but perform selective marking based on the source rate value inserted in each packet, We propose a new strategy for source rate encoding in forward packets, directly applicable to existing network protocols (e.g. IP). The scheme supports differentiated classes with respect to rate allocation. We test, by simulation, this encoding mechanism as well as the performance of the router and source algorithms. (C) 2002 Elsevier Science B.V. All rights reserved.
Cloud-native applications add a layer of abstraction to the underlying distributed computing system, defining a high-level, self-scaling and self-managed architecture of different microservices linked by a messaging b...
详细信息
ISBN:
(纸本)9783319992532;9783319992525
Cloud-native applications add a layer of abstraction to the underlying distributed computing system, defining a high-level, self-scaling and self-managed architecture of different microservices linked by a messaging bus. Creating new algorithms that tap these architectural patterns and at the same time employ distributed resources efficiently is a challenge we will be taking up in this paper. We introduce KafkEO, a cloud-native evolutionary algorithms framework that is prepared to work with different implementations of evolutionary algorithms and other population-based metaheuristics by using micro-populations and stateless services as the main building blocks;KafkEO is an attempt to map the traditional evolutionary algorithm to this new cloud-native format. As far as we know, this is the first architecture of this kind that has been published and tested, and is free software and vendor-independent, based on OpenWhisk and Kafka. This paper presents a proof of concept, examines its cost, and tests the impact on the algorithm of the design around cloud-native and asynchronous system by comparing it on the well known BBOB benchmarks with other pool-based architectures, with which it has a remarkable functional resemblance. KafkEO results are quite competitive with similar architectures.
暂无评论