Effect of Batching on Performance Characteristics of DCOM-based Distributed Implementation of Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.7615

@InProceedings{tanev:2000:PETTA,

• author = "Ivan Tanev and Takashi Uozumi and Koichi Ono",
• title = "Effect of Batching on Performance Characteristics of DCOM-based Distributed Implementation of Genetic Programming",
• booktitle = "Proceedings of International Conference on Performance Evaluation (PerETTA-2000)",
• year = "2000",
• pages = "32--37",
• keywords = "genetic algorithms, genetic programming",
• abstract = "performance characteristics of developed distributed implementation of genetic programming (DIGP). Proposed implementation exploits the inherent medium-grained parallelism among evaluation of the individuals in genetic programming. It runs on cost-efficient configurations of clusters in LAN and/or Internet environment. DIGP exploits distributed component object model (DCOM) as an underlying host to host communication paradigm, which offers such advantages as binary standardisation, platform-, machine- and protocol-neutrality and seamless integration with different Internet protocols. In addition, as a true system model, DCOM offers generic support for the issues of naming, locating and protecting the distributed entities in proposed implementation. However, as an application-level protocol, DCOM features significant software overheads, which, in some cases might be above the cost of distributed computations. We introduce the approach of batching for reducing the specific software overhead of DCOM, and present the analytically estimated and empirically proved effect on the performance characteristics of developed DIGP.",

}

Genetic Programming entries for Ivan T Tanev Takashi Uozumi Koichi Ono

Citations