SELF LEARNING AND EFFICIENT HEALTH STATUS

ANALYSIS FOR A CORE ROUTER SYSTEM

SUBMITTED BY:
MR. SANDEEP REDDY(16C01A0536)
K. KISHAN (16C01A0530)
N. NIKITHA(16C01A0536)

GUIDE:
Mrs. M. Manasa(Asst Professor)
CONTENTS
1. Abstract
2. Introduction
3. Existing System and its disadvantages
4. Proposed System and its advantages
5. System Requirements
6. Modules
7. System Design
8. Output Screens
9. Conclusion
ABSTRACT
⇢ The health status of core router systems needs to be analyzed efficiently
in order to ensure high reliability and timely error recovery.
⇢ We present an iterative self learning procedure for assessing the health
status of a core router.
⇢ This procedure first computes a representative feature matrix to capture different
characteristics of time-series data.
⇢ Moreover, bothminimum-redundancy-maximum-relevance (mRMR) method and fully-
connected feed forward autoencoder are applied to further reduce dimensionality of
extracted feature matrix.
⇢ Field data collected from a set of commercial core routers are used to
experimentally validate the proposed health-status analyzer.
INTRODUCTION
⇢ The network devices (such as routers) used in the core network are
complex hardware/software systems that are vulnerable to hard-to
detect hard-to-recover errors.
⇢ Consider a multi-card chassis system, which is a widely used architecture in core routers.A
srange of failures can occur in such a complex system.
⇢ Traditional reactive fault tolerance aims at repair after failures occur.
⇢ Proactive fault tolerance is more promising because it takes preventive action before a failure
occurs.
⇢ The effectiveness of proactive fault-tolerance solutions depends on
whether the health status of core router can be accurately identified
in a timely manner
⇢ Health assessment is an essential component in a data driven proactive
fault-tolerant scheme.
⇢ Data-driven techniques are especially promising for a core router that uses monitors to log a
large amount of operational data from different
functional units.
⇢ Therefore, operational data are collected in the form of time series where data points are
indexed in time order.
⇢ A common way to identify a system’s health status is to feed its
features to an anomaly detector to see whether any data points are
statistical outliers.
EXISTING SYSTEM
⇢ The network devices (such as routers) used in the core network are
complex hardware/software systems such as multi-card chassis system
can have tens of separate cards ,and each card can have hundreds of
components that are vulnerable to hard-to detect/hard-to-recover errors .

Disadvantages
• This system can lead to multiple hardware and software failures.
• Whenever a hardware or Software failure occurs, it is difficult for debug
technicians to accurately identify the root cause of this failure and take
effective corrective actions.
PROPOSED SYSTEM
⇢ We have presented a self-learning health analyzer for a complex core
router system.
⇢ The Effectiveness of proactive fault-tolerance solutions depends on whether the
health status of core routers can be accurately identified in a timely manner.
⇢ Therefore,in this paper, we present the design of an efficient self-learning health
analyzer that can be applied to a commercial core router system.
Advantages
• Therefore by this system we can predict any hardware and software failure before hand.
• This system also helps us to identify the location where failure occured and helps us to
resolve easily.
HARDWARE REQUIREMENTS
⇢ System :Pentium !V 2.4 GHZ.
⇢ Hard Disk :40 GB
⇢ Monitor :14' Colour Monitor
⇢ Mouse :Optical Mouse
⇢ Ram :4 GB
SOFTWARE REQUIREMENTS
⇢ Operating system :Windows/windows 10.
⇢ Coding Language :Python
⇢ Front-End :Python
⇢ Designing :Html, css, javascript
⇢ Database :MySQL.
MODULES
⇢ [Link]
⇢ [Link]
⇢ [Link]
⇢ [Link] Learning
User
⇢ The core network, also referred to as the network backbone, is
responsible for the transfer of a large amount of traffic.
⇢ In this users are going to send messages to their [Link]
send the messages by using sentbox and receive the messages using
inbox .
Router
⇢ The health status of core router systems needs to be analyzed efficiently
in order to ensure high reliability and timely error recovery.
⇢ The model can be trained on the basis of routers ,user can
see that the logs not only extract values of different features, but
also record the corresponding time stamps.
⇢ Therefore, operational data are collected in the form of time series
where data points are indexed in time order.
Admin
⇢ Admin gives the activation permission to users , admin
can verify the router status of uses and their CPU usage total ram
used ram storage and the router status.
⇢ The admin can calculate raw SVM support vector machine. the
admin can also calculate the raw svm DT (decision tree algorithm).
Machine Learning
⇢ Machine learning has been extensively used to benefit router healthcare.
⇢ Machine Learning can be used in identification of router, give
personalized suggestion to Users.
⇢ Many techniques have been done in this area which include SVM
classifier,Machine-learning techniques, such as support vector machine
(SVM), decision tree (DT), and artificial neural network, can be used as
the base classifier.
⇢ The machine learning models can be trained to predict future outcomes
Neural Network with SVM was used to predict outbreaks State.
SYSTEM ARCHITECTURE
User DFD
Router DFD
Admin DFD
UML Diagrams
User Use case diagram
Router Use Case Diagram
Admin Use Case Diagram
Class Diagram
Sequence Diagram
Activity Diagram
SYSTEM
Types of tests
TESTING ⇢ Unit testing
• The purpose of testing is to discover errors. ⇢ Integration testing
• Testing is the process of trying to discover ⇢ Functional testing
every conceivable fault or weakness in a
work ⇢ System test
Product.
⇢ White box testing
• It provides a way to check the functionality
of components,sub assemblies, assemblies
and/or a finished product It is the process of
exercising software with the intent of
ensuring that the Software system meets its
requirements and user experience
OUTPUT SCREENS
Home Page
User register
User Login
User Home Page
Admin Home Page
Admin SVM
Admin RAW DT
Router Home Page
CONCLUSION
⇢ We have presented a feature-based self-learning health analyzer for a
complex core router system.
⇢ First, both the statistical-modeling-based feature extraction and auto
encoder based feature learning have been utilized to capture different
characteristics of time-series data.
⇢ Next, in the self-learning framework, the model for health analysis is
iteratively updated using both labeled and unlabeled data. The
effectiveness of the health analyzer has been validated using a
comprehensive set of field data collected from a set of commercial core
routers.
THANK
YOU!!!!!