4 Ppt on YARN MapReduce 31 10 20
4 Ppt on YARN MapReduce 31 10 20
4 Ppt on YARN MapReduce 31 10 20
Unit IV
Topic: working of YARN in Hadoop
July-Dec
2
Prerequisite of Course
3
Course Objective
•To introduce various techniques for representation of the data in the real
world.
4
Architecture of MapReduce
5
Working of MapReduce:
6
Role of Job tracker:-
7
Role of task tracker:-
8
Architecture of YARN:-
4 core components of –
1. Resource manager- only responsible for resource allocation.
2. Node manager- responsible for launching the application
container .
3. Application Master-
When the container has errors ,the application master will apply for
another resource from the scheduler to continue the process.
4. Container- it is the source unit which is available node splitting the
organization resources.
9
YARN
YARN:-
It means yet
Another
Resource
Negotiator.
10
YARN
It is also called next generation MapReduce.
Pupose of YARN:-
1. Divide the tasks for the Jobtracker.
2. Resources are managed by Resource manager.
3. Jobs are traced by application master.
4. Task tracker has become the node manager.
5. So global resource manager and local ode manger compose the
data computing framework.
6. Resource manager will be the resources distributor while the
apllication master is response for the communication with resource
manger and cooprate the node manager to complete the tasks.
11
Advantages of YARN:-
12
Uses of MapReduce:-
1.At google –
• index building for google search
• Article clustering for google news
• Statistical machine translation
2. At yahoo-
• Spam detection for yahoo mail
• Index building for yahoo search
3.At facebook-
• Ad optimization
• Spam detection
13
Research Scope
• Role of Data analytics in multiple disciplines
14
References
15
References
3. Barlow, M., Real-Time Big Data Analytics: Emerging
Architecture, O Reilly, 2013
4. Schonberger, V.M. , Kenneth Cukier, K., Big Data, John
Murray Publishers, 2013,
16
17