Multi Controller Load Balancing in Software Defined Networks
Title:- AI-MLB: A Triple Tier Framework for Intellectual Edge Assisted Multicontroller Load Balancing in SDN using NFV and Artificial Intelligence Algorithms
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Implementation Plan:
---------------------------------
Step 1: We create a Network, it consists of 50- Iot Users, 1- Gateway, 1- Global Controller, 1- Virtual Switch, 1- Virtual Controller and 4- Local Controller.
Step 2: Next, we perform the User Selection process, in this process we realized by fixing the minimum threshold based on device efficiency (battery level) and bandwidth as parameters using Interval Type-II Hesitant Fuzzy Entropy Measure (IT-II HFE).
Step 3: Next, Flow Classification & Prioritization using Deep Learning process, In this process the arrived flows are classified based on sensitivity. For classification we used Non-Deep Lightweight Parallel Network (ND-LPN).
Step 4: Next, perform the Hybrid Multi-Controller Load Balancing (MCLB), Here 2 significant load balancing approaches termed,
4.1: Proactive Load Balancing (PLB), Here, the flow engineer predicts the future loads of every controller based on their load history and current load (sum of the requests received and packet in messages) by incorporating a Dual Agent-based Geometric Actor-Critic Algorithm (DAGAC).
4.2: Reactive Load Balancing (RLB), Here the lightly loaded controller is optimally chosen by the global controller for scattering the burden on a single controller. For this, a Hybrid Leader Optimization Algorithm (HLO) is used.
Step 5: The proposed work’s efficiency can be measured through several performance metrics as enumerated below,
5.1 Controller Throughput (kpbs)Vs Total Load
5.2 Incoming Load Vs Average Packet Loss (%)
5.3 Incoming Load Vs Average Response Time (ms)
5.4 Incoming Load Vs Average Number of Migration (ms)
5.5 Network Load Vs. Number of Arrived Users
==============================================================================================
Software Requirements:
------------------------------------------
1) Ns-3.26
3) Ubuntu 14.04 [32-bit]
============================================================================================================
Note:-
---------
We perform the EXISTING process based on the REFERENCE Title 4: - An Improved Switch Migration Decision Algorithm for SDN Load Balancing
Execution Steps:
-----------------------------
Open the ternimal (ctrl + alt + t)
change the directory on terminal.
First execute the files
change the image locations in the main files.
command to execute :
./waf --run filename --vis
Note:
------------
1) use our module else you get error.
2)Refer how to add our model.txt to add module
=========================================================================================
This video aims to provide guidance towards the resolution of your inquiry
regarding the provision of assistance in selecting a topic for cloud computing
studies.
We have an experienced research team for your cloud computing research
study.
For Research Support contact us:
Our website: https://cloudcomputingprojects.net/
Mail us at: [email protected]
Call us at : +91 94448 69228