Title
Scalability and Reliability of Port Simulation Workflow on Slurm
Authors
[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ P ] employee | [ D ] phd student
Scientific discipline (Law 2.0)
Year of publication
2025
Chapter type
chapter in monograph / paper
Publication language
english
Keywords
EN
- Port simulation
- Berth allocation problem
- Stochastic optimization
- Slurm
- Reliability
- Scalability
Abstract
EN Parallelizing of container terminal seaside layout optimizer using a Slurm cluster is considered in this paper. Maritime container terminals have quays divided into discrete berth segments. The number of berths and their lengths have to be adjusted to the arriving vessel traffic for high quality of service. A particular partition of the terminal quay into berths is called terminal layout. The vessel traffic is represented by the stochastic ship traffic model (STM). An instantiation of the STM is a specific ship arrival scenario. Evaluation of a quay layout on an arrival scenario requires scheduling vessels on the berths, which is a classic Berth Allocation Problem (BAP) of maritime logistics. The problem of partitioning terminal quay into berths, subject to vessel traffic, for high quality of service will be called a Stochastic Quay Partitioning Problem (SQPP). In this paper, we present a hill-climber metaheuristic for SQPP. The process of the partition evaluation is conducted on a Slurm cluster by a simple workflow. We report on the scalability and reliability of this workflow.
Date of online publication
01.04.2025
Pages (from - to)
91 - 102
Ministry points / chapter
20
Ministry points / conference (CORE)
20
System created by Poznań University of Technology
and Poznan Supercomputing and Networking Center
Log in through eKonto to add to SIS