# American Institute of Mathematical Sciences

March  2018, 8(1): 47-62. doi: 10.3934/naco.2018003

## Globalizer: A novel supercomputer software system for solving time-consuming global optimization problems

 Lobachevsky State University of Nizhni Novgorod, 23, Gagarin Avenue, Nizhni Novgorod, Russia

* Corresponding author: Konstantin Barkalov

Received  December 2016 Revised  September 2017 Published  March 2018

Fund Project: The study was supported by the Russian Science Foundation, project No 16-11-10150.

In this paper, we describe the Globalizer software system for solving the global optimization problems. The system is designed to maximize the use of computational potential of the modern high-performance computational systems in order to solve the most time-consuming optimization problems. The highly parallel computations are facilitated using various distinctive computational schemes: processing several optimization iterations simultaneously, reducing multidimensional optimization problems using multiple Peano space-filling curves, and multi-stage computing based on the nested block reduction schemes. These novelties provide for the use of the supercomputer system capabilities with shared and distributed memory and with large numbers of processors to solve the global optimization problems efficiently.

Citation: Victor Gergel, Konstantin Barkalov, Alexander Sysoyev. Globalizer: A novel supercomputer software system for solving time-consuming global optimization problems. Numerical Algebra, Control & Optimization, 2018, 8 (1) : 47-62. doi: 10.3934/naco.2018003
##### References:

show all references

##### References:
A Peano curve approximation for the third density level
The computational scheme for obtaining the value of the reduced one-dimensional function $\varphi(y(x))$
The architecture of the Globalizer system
Averaged number of executed iterations of the compared global optimization methods
 $N$ Problem class DIRECT DIRECTl MAGS 4 Simple $>$47282(4) 18983 11953 Hard $>$95708(7) 68754 25263 5 Simple $>$16057(1) 16758 15920 Hard $>$217215(16) $>$269064(4) $>$148342(4)
 $N$ Problem class DIRECT DIRECTl MAGS 4 Simple $>$47282(4) 18983 11953 Hard $>$95708(7) 68754 25263 5 Simple $>$16057(1) 16758 15920 Hard $>$217215(16) $>$269064(4) $>$148342(4)
Averaged numbers of iterations executed by GPMAGS for solving the test optimization problems
 p $N=4$ $N=5$ Simple Hard Simple Hard Ⅰ Serial trial computations 1 11953 25263 15920 $>$148342 (4) Ⅱ Parallel computations 2 4762 11178 13378 109075 on CPU 4 2372 5972 5203 51868 8 1393 2874 3773 51868 Ⅲ Parallel computations 60 171 393 382 3452 on Xeon Phi 120 85 182 249 1306 240 42 103 97 381
 p $N=4$ $N=5$ Simple Hard Simple Hard Ⅰ Serial trial computations 1 11953 25263 15920 $>$148342 (4) Ⅱ Parallel computations 2 4762 11178 13378 109075 on CPU 4 2372 5972 5203 51868 8 1393 2874 3773 51868 Ⅲ Parallel computations 60 171 393 382 3452 on Xeon Phi 120 85 182 249 1306 240 42 103 97 381
Speedup of parallel computations executed by GPMAGS
 p $N=4$ $N=5$ Simple Hard Simple Hard Ⅰ Serial trial computations. 1 11953 25263 15920 > 148342 (4) Average number of iterations Ⅱ Parallel computations 2 2.52 2.32 1.21 1.41 of CPU. 4 5.05 4.24 3.13 2.92 Speedup 8 8.68 8.88 4.24 6.66 Ⅲ Parallel computations 60 8.18 7.37 9.99 6.66 of Xeon Phi. 120 16.316 15.815 15.215 17.317 Speedup 240 33.133 27.827 38.838 59.359
 p $N=4$ $N=5$ Simple Hard Simple Hard Ⅰ Serial trial computations. 1 11953 25263 15920 > 148342 (4) Average number of iterations Ⅱ Parallel computations 2 2.52 2.32 1.21 1.41 of CPU. 4 5.05 4.24 3.13 2.92 Speedup 8 8.68 8.88 4.24 6.66 Ⅲ Parallel computations 60 8.18 7.37 9.99 6.66 of Xeon Phi. 120 16.316 15.815 15.215 17.317 Speedup 240 33.133 27.827 38.838 59.359
 [1] Felix Finster, Jürg Fröhlich, Marco Oppio, Claudio F. Paganini. Causal fermion systems and the ETH approach to quantum theory. Discrete & Continuous Dynamical Systems - S, 2020  doi: 10.3934/dcdss.2020451 [2] Maoding Zhen, Binlin Zhang, Vicenţiu D. Rădulescu. Normalized solutions for nonlinear coupled fractional systems: Low and high perturbations in the attractive case. Discrete & Continuous Dynamical Systems - A, 2020  doi: 10.3934/dcds.2020379 [3] Nicolas Rougerie. On two properties of the Fisher information. Kinetic & Related Models, , () : -. doi: 10.3934/krm.2020049 [4] Vieri Benci, Marco Cococcioni. The algorithmic numbers in non-archimedean numerical computing environments. Discrete & Continuous Dynamical Systems - S, 2020  doi: 10.3934/dcdss.2020449 [5] Shengxin Zhu, Tongxiang Gu, Xingping Liu. AIMS: Average information matrix splitting. Mathematical Foundations of Computing, 2020, 3 (4) : 301-308. doi: 10.3934/mfc.2020012 [6] Predrag S. Stanimirović, Branislav Ivanov, Haifeng Ma, Dijana Mosić. A survey of gradient methods for solving nonlinear optimization. Electronic Research Archive, 2020, 28 (4) : 1573-1624. doi: 10.3934/era.2020115 [7] Haiyu Liu, Rongmin Zhu, Yuxian Geng. Gorenstein global dimensions relative to balanced pairs. Electronic Research Archive, 2020, 28 (4) : 1563-1571. doi: 10.3934/era.2020082 [8] Jianhua Huang, Yanbin Tang, Ming Wang. Singular support of the global attractor for a damped BBM equation. Discrete & Continuous Dynamical Systems - B, 2020  doi: 10.3934/dcdsb.2020345 [9] Yueyang Zheng, Jingtao Shi. A stackelberg game of backward stochastic differential equations with partial information. Mathematical Control & Related Fields, 2020  doi: 10.3934/mcrf.2020047 [10] Pierre-Etienne Druet. A theory of generalised solutions for ideal gas mixtures with Maxwell-Stefan diffusion. Discrete & Continuous Dynamical Systems - S, 2020  doi: 10.3934/dcdss.2020458 [11] Juan Pablo Pinasco, Mauro Rodriguez Cartabia, Nicolas Saintier. Evolutionary game theory in mixed strategies: From microscopic interactions to kinetic equations. Kinetic & Related Models, , () : -. doi: 10.3934/krm.2020051 [12] Youming Guo, Tingting Li. Optimal control strategies for an online game addiction model with low and high risk exposure. Discrete & Continuous Dynamical Systems - B, 2020  doi: 10.3934/dcdsb.2020347 [13] Alberto Bressan, Sondre Tesdal Galtung. A 2-dimensional shape optimization problem for tree branches. Networks & Heterogeneous Media, 2020  doi: 10.3934/nhm.2020031 [14] Cheng He, Changzheng Qu. Global weak solutions for the two-component Novikov equation. Electronic Research Archive, 2020, 28 (4) : 1545-1562. doi: 10.3934/era.2020081 [15] Claudianor O. Alves, Rodrigo C. M. Nemer, Sergio H. Monari Soares. The use of the Morse theory to estimate the number of nontrivial solutions of a nonlinear Schrödinger equation with a magnetic field. Communications on Pure & Applied Analysis, , () : -. doi: 10.3934/cpaa.2020276 [16] Peizhao Yu, Guoshan Zhang, Yi Zhang. Decoupling of cubic polynomial matrix systems. Numerical Algebra, Control & Optimization, 2021, 11 (1) : 13-26. doi: 10.3934/naco.2020012 [17] M. S. Lee, H. G. Harno, B. S. Goh, K. H. Lim. On the bang-bang control approach via a component-wise line search strategy for unconstrained optimization. Numerical Algebra, Control & Optimization, 2021, 11 (1) : 45-61. doi: 10.3934/naco.2020014 [18] Mohammed Abdulrazaq Kahya, Suhaib Abduljabbar Altamir, Zakariya Yahya Algamal. Improving whale optimization algorithm for feature selection with a time-varying transfer function. Numerical Algebra, Control & Optimization, 2021, 11 (1) : 87-98. doi: 10.3934/naco.2020017 [19] Ahmad Z. Fino, Wenhui Chen. A global existence result for two-dimensional semilinear strongly damped wave equation with mixed nonlinearity in an exterior domain. Communications on Pure & Applied Analysis, 2020, 19 (12) : 5387-5411. doi: 10.3934/cpaa.2020243 [20] Mengni Li. Global regularity for a class of Monge-Ampère type equations with nonzero boundary conditions. Communications on Pure & Applied Analysis, , () : -. doi: 10.3934/cpaa.2020267

Impact Factor: