US 12,190,162 B2
Optimized allocation of resources to nodes of an automation system based on monte carlo simulation
Andrés Botero Halblaub, Kirchseeon (DE); and Jan Richter, Lauf a. d. Pengnitz (DE)
Assigned to Siemens Aktiengesellschaft, Munich (DE)
Appl. No. 17/041,928
Filed by Siemens Aktiengesellschaft, Munich (DE)
PCT Filed Feb. 28, 2019, PCT No. PCT/EP2019/054959
§ 371(c)(1), (2) Date Sep. 25, 2020,
PCT Pub. No. WO2019/185278, PCT Pub. Date Oct. 3, 2019.
Claims priority of application No. 18164275 (EP), filed on Mar. 27, 2018.
Prior Publication US 2021/0141673 A1, May 13, 2021
Int. Cl. G06F 9/50 (2006.01); G06F 9/52 (2006.01); G06F 30/20 (2020.01); G06F 111/08 (2020.01)
CPC G06F 9/5038 (2013.01) [G06F 9/5055 (2013.01); G06F 9/52 (2013.01); G06F 30/20 (2020.01); G06F 2111/08 (2020.01)] 15 Claims
OG exemplary drawing
 
1. A method for computer-implemented configuration of an automation system by means of a processing unit of a computer system, the automation system having computing nodes configured to allocate resources to a software system in the computer system, the method comprising:
a) receiving, at the processing unit, a computer generated data set of sequencing constraints describing the software system to be executed on the automation system and comprising software components and runnable function entities distributed over the computing nodes as well as timing, sequencing and concurrency information of the runnable function entities;
b) generating, by the processing unit, a transition matrix from the computer generated data set of sequencing constraints, the transition matrix having a plurality of matrix elements each describing, by a transition value, a transition from a runnable function entity to another runnable function entity;
c) receiving, at the processing unit, a computer generated communication matrix describing communication links between the computing nodes in the automation system;
d) generating, by the processing unit, a Markov chain out of the computer generated data set of sequencing constraints and the computer generated communication matrix;
e) generating, by the processing unit, a distribution function from the Markov chain describing, for a specific timing distribution of the software components and runnable function entities, resources of the computing nodes to be allocated to the software components and the runnable function entities;
f) performing, by the processing unit, Monte Carlo simulations for different resources used by the runnable function entities and different execution times of the runnable function entities to output a graph of resources needed by the runnable function entities and to determine an optimum allocation of resources for the runnable function entities;
g) optimally allocating, by the processing unit in accordance with the optimum allocation of resources, the resources to the runnable function entities of the software system; and
h) executing, by the processing unit, the runnable function entities.