Storage area network simulation

The research on Storage Area Network has been performed for more than 10 years. Though the complexity and cost has been dropping, defect implementation of Storage Area Network will cause a huge amount of losses. Therefore, proper design and testing are needed before Storage Area Network is applied i...

Full description

Saved in:
Bibliographic Details
Main Author: Rudy.
Other Authors: Foh Chuan Heng
Format: Final Year Project
Language:English
Published: 2009
Subjects:
Online Access:http://hdl.handle.net/10356/16914
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:The research on Storage Area Network has been performed for more than 10 years. Though the complexity and cost has been dropping, defect implementation of Storage Area Network will cause a huge amount of losses. Therefore, proper design and testing are needed before Storage Area Network is applied in the real world application. Various Simulators are available for the research and experiment on Storage Area Network. In this project, we used SimSAN as our simulation tool because of its specialization in Storage Area Network and implementation of Fibre Channel Protoco. SimSAN is built in OMNET++ framework (open-source network simulation environment), hence it gives the flexibility of creating our own topology. On the analysis of the output files generated from SimSAN, we discovered that for higher IO size network, we should get less number of frames being transferred in the network and hence, the CPU utilization and IO request is less. For more hosts (with the same IO size), we should get more frame being transferred, higher CPU utilization on the Switch and Device (CPU utilization on the Host remains the same), and higher IO request. Customized topology is made available by editing the NED (Network Description Language from OMNET++) files; compile it to get the source code and make an executable file out of it. This executable file is used to run the simulation. Taking the advantage of the flexibility, we conclude that the possibility of research on the best topology for Storage Area Network is not impossible.