Genetic local search and hardness of approximation for the server load balancing problem
- Авторы: Kochetov Y.A.1,2, Panin A.A.1,2, Plyasunov A.V.1,2
-
Учреждения:
- Novosibirsk State University
- Sobolev Institute of Mathematics, Siberian Branch
- Выпуск: Том 78, № 3 (2017)
- Страницы: 425-434
- Раздел: Stochastic Systems, Queueing Systems
- URL: https://ogarev-online.ru/0005-1179/article/view/150555
- DOI: https://doi.org/10.1134/S0005117917030043
- ID: 150555
Цитировать
Аннотация
We consider a well-known NP-hard server load balancing problem. We study the computational complexity of finding approximate solutions with guaranteed accuracy estimate. We show that this problem is Log-APX-hard with respect to PTAS reductions. To solve the problem, we develop an approximate method based on the ideas of genetic local search. We show results of computational experiments.
Ключевые слова
Об авторах
Yu. Kochetov
Novosibirsk State University; Sobolev Institute of Mathematics, Siberian Branch
Автор, ответственный за переписку.
Email: jkochet@math.nsc.ru
Россия, Novosibirsk; Novosibirsk
A. Panin
Novosibirsk State University; Sobolev Institute of Mathematics, Siberian Branch
Email: jkochet@math.nsc.ru
Россия, Novosibirsk; Novosibirsk
A. Plyasunov
Novosibirsk State University; Sobolev Institute of Mathematics, Siberian Branch
Email: jkochet@math.nsc.ru
Россия, Novosibirsk; Novosibirsk
Дополнительные файлы
