A Simulation Methodology for Distributed Storage

Peter Sobe; Moritz Baum; Sergej Heckel; Janine Krüger

A Simulation Methodology for Distributed Storage

Peter Sobe, Moritz Baum, Sergej Heckel, Janine Krüger

Institute of Computer Engineering

University of Greifswald

Abstract

An event-based simulation system is introduced, designed for performance assessment of distributed storage systems. Such storage systems employ the commu- nication network to connect several distributed storage units to a virtual storage system that provides higher capacity, better performance and that is tolerant against storage unit failures. By simulation, several data distribution st rategies and redundancy lay- outs can be analyzed, particularly focusing on effects caused by load imbalances and asynchronous operation of storage units. In this paper, a methodology is introduced to simulate (i) the effects of scaling the distribution degree and (ii) the effects of access load balancing techniques. This helps to design proper data and redundancy layouts for distributed data storage systems.

Original language	English
Journal	BALTIC CONFERENCE on Advanced Topics in Telecommunications
Pages (from-to)	7-16
Number of pages	10
Publication status	Published - 01.03.2019

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Cite this

@article{bc02859a7ba9474e8050fbb801792b83,

title = "A Simulation Methodology for Distributed Storage",

abstract = "An event-based simulation system is introduced, designed for performance assessment of distributed storage systems. Such storage systems employ the commu- nication network to connect several distributed storage units to a virtual storage system that provides higher capacity, better performance and that is tolerant against storage unit failures. By simulation, several data distribution st rategies and redundancy lay- outs can be analyzed, particularly focusing on effects caused by load imbalances and asynchronous operation of storage units. In this paper, a methodology is introduced to simulate (i) the effects of scaling the distribution degree and (ii) the effects of access load balancing techniques. This helps to design proper data and redundancy layouts for distributed data storage systems.",

author = "Peter Sobe and Moritz Baum and Sergej Heckel and Janine Kr{\"u}ger",

year = "2019",

month = mar,

day = "1",

language = "English",

pages = "7--16",

journal = "BALTIC CONFERENCE on Advanced Topics in Telecommunications",

}

TY - JOUR

T1 - A Simulation Methodology for Distributed Storage

AU - Sobe, Peter

AU - Baum, Moritz

AU - Heckel, Sergej

AU - Krüger, Janine

PY - 2019/3/1

Y1 - 2019/3/1

N2 - An event-based simulation system is introduced, designed for performance assessment of distributed storage systems. Such storage systems employ the commu- nication network to connect several distributed storage units to a virtual storage system that provides higher capacity, better performance and that is tolerant against storage unit failures. By simulation, several data distribution st rategies and redundancy lay- outs can be analyzed, particularly focusing on effects caused by load imbalances and asynchronous operation of storage units. In this paper, a methodology is introduced to simulate (i) the effects of scaling the distribution degree and (ii) the effects of access load balancing techniques. This helps to design proper data and redundancy layouts for distributed data storage systems.

AB - An event-based simulation system is introduced, designed for performance assessment of distributed storage systems. Such storage systems employ the commu- nication network to connect several distributed storage units to a virtual storage system that provides higher capacity, better performance and that is tolerant against storage unit failures. By simulation, several data distribution st rategies and redundancy lay- outs can be analyzed, particularly focusing on effects caused by load imbalances and asynchronous operation of storage units. In this paper, a methodology is introduced to simulate (i) the effects of scaling the distribution degree and (ii) the effects of access load balancing techniques. This helps to design proper data and redundancy layouts for distributed data storage systems.

UR - https://www.researchgate.net/publication/242084657_A_Simulation_Methodology_for_Distributed_Storage

M3 - Journal articles

SP - 7

EP - 16

JO - BALTIC CONFERENCE on Advanced Topics in Telecommunications

JF - BALTIC CONFERENCE on Advanced Topics in Telecommunications

ER -

A Simulation Methodology for Distributed Storage

Abstract

UN SDGs

Other files and links

Fingerprint

Cite this