www.edureka.co/big-data-and-hadoop
Introduction to big data and hadoop
Slide 2 www.edureka.co/big-data-and-hadoop
Slide 3 www.edureka.co/big-data-and-hadoop
At the end of this session , you will understand the:
→ Big Data Learning Paths
→ Big Data Introduction
→ Hadoop and Its Eco-System
→ Hadoop Architecture
→ Next Step on How to Setup Hadoop
Slide 4 www.edureka.co/big-data-and-hadoop
• Java / Python / Ruby
• Hadoop Eco-system
• NoSQL DB
• Spark
• Linux Administration
• Cluster Management
• Cluster Performance
• Virtualization
• Statistics Skills
• Machine Learning
• Hadoop Essentials
• Expertise in R
Developer/Testing
Administration
Data Analyst
Big Data and Hadoop
MapReduce
Design Patterns
Apache
Spark & Scala
Apache Cassandra
Linux Administration Hadoop Administration
Data Science
Business Analytics
Using R
Advance Predictive
Modelling in R
Talend for Big Data
Data Visualization
Using Tableau
Slide 5 www.edureka.co/big-data-and-hadoop
→ Lots of Data (Terabytes or Petabytes)
→ Big data is the term for a collection of data sets so
large and complex that it becomes difficult to process
using on-hand database management tools or
traditional data processing applications
→ The challenges include capture, curation, storage,
search, sharing, transfer, analysis, and visualization Big Data
Slide 6 www.edureka.co/big-data-and-hadoop
→ Systems / Enterprises generate huge amount of data from Terabytes to Petabytes of information
Stock market generates about one terabyte of new trade data per day to
perform stock trading analytics to determine trends for optimal trades
Slide 7 www.edureka.co/big-data-and-hadoop
→ By 2020, IDC (International Data Corporation) predicts the number will have reached 40,000 EB, or 40 Zettabytes (ZB)
→ The world’s information is doubling every two years. By 2020 the world will generate 50 times the amount of
information and 75 times the number of information containers
Slide 8 www.edureka.co/big-data-and-hadoop
IBM’s Definition – Big Data Characteristics
http://www-01.ibm.com/software/data/bigdata/
VOLUME
Web
logs
Images
Videos
Audios
Sensor
Data
VARIET
Y
VELOCITY VERACITY
Min Max Mean SD
4.3 7.9 5.84 0.83
2.0 4.4 3.05 0.43
0.1 2.5 1.20 0.76
Slide 9 www.edureka.co/big-data-and-hadoop
Hello There!!
My name is Annie.
I love quizzes and
puzzles and I am here to
make you guys think and
answer my questions.
Slide 10 www.edureka.co/big-data-and-hadoop
Map the following to corresponding data type:
» XML files, e-mail body
» Audio, Video, Images, Archived documents
» Data from Enterprise systems (ERP, CRM etc.)
Slide 11 www.edureka.co/big-data-and-hadoop
Ans. XML files, e-mail body → Semi-structured data
Audio, Video, Image, Files, Archived documents → Unstructured data
Data from Enterprise systems (ERP, CRM etc.) → Structured data
Slide 12 www.edureka.co/big-data-and-hadoop
More on Big Data
• http://www.edureka.in/blog/the-hype-behind-big-data/
Why Hadoop?
• http://www.edureka.in/blog/why-hadoop/
Opportunities in Hadoop
• http://www.edureka.in/blog/jobs-in-hadoop/
Big Data
• http://en.wikipedia.org/wiki/Big_Data
IBM’s definition – Big Data Characteristics
• http://www-01.ibm.com/software/data/bigdata/
Slide 13Slide 13Slide 13 www.edureka.co/big-data-and-hadoop
→ Web and e-tailing
» Recommendation Engines
» Ad Targeting
» Search Quality
» Abuse and Click Fraud Detection
→ Telecommunications
» Customer Churn Prevention
» Network Performance Optimization
» Calling Data Record (CDR) Analysis
» Analysing Network to Predict Failure
http://wiki.apache.org/hadoop/PoweredBy
Slide 14Slide 14Slide 14 www.edureka.co/big-data-and-hadoop
→ Government
» Fraud Detection and Cyber Security
» Welfare Schemes
» Justice
→ Healthcare and Life Sciences
» Health Information Exchange
» Gene Sequencing
» Serialization
» Healthcare Service Quality Improvements
» Drug Safety
http://wiki.apache.org/hadoop/PoweredBy
Slide 15Slide 15Slide 15 www.edureka.co/big-data-and-hadoop
→ Banks and Financial services
» Modeling True Risk
» Threat Analysis
» Fraud Detection
» Trade Surveillance
» Credit Scoring and Analysis
→ Retail
» Point of Sales Transaction Analysis
» Customer Churn Analysis
» Sentiment Analysis
http://wiki.apache.org/hadoop/PoweredBy
Slide 16Slide 16Slide 16 www.edureka.co/big-data-and-hadoop
→ Insight into data can provide Business Advantage.
→ Some key early indicators can mean Fortunes to Business.
→ More Precise Analysis with more data.
*Sears was using traditional systems such as Oracle Exadata, Teradata and SAS etc., to store and process the customer activity and sales data.
Case Study: Sears Holding Corporation
Slide 17Slide 17Slide 17 www.edureka.co/big-data-and-hadoop
Mostly Append
BI Reports + Interactive Apps
RDBMS (Aggregated Data)
ETL Compute Grid
Storage only Grid (Original Raw Data)
Collection
Inctrumentation
A meagre
10% of the
~2PB data is
available for
BI
Storage
2. Moving data to compute doesn’
t scale
90% of
the ~2PB
archived
Processing
3. Premature data
death
1. Can’t explore original
high fidelity raw data
Slide 18Slide 18Slide 18 www.edureka.co/big-data-and-hadoop
Mostly Append
BI Reports + Interactive Apps
RDBMS (Aggregated Data)
Hadoop : Storage + Compute Grid
Collection
Instrumentation
Both
Storage
And
Processing
Entire ~2PB
Data is
available for
processing
No Data
Archiving
1. Data Exploration &
Advanced analytics
2. Scalable throughput for ETL &
aggregation
3. Keep data alive
forever
*Sears moved to a 300-Node Hadoop cluster to keep 100% of its data available for processing rather than a meagre 10% as
was the case with existing Non-Hadoop solutions.
Slide 19Slide 19Slide 19 www.edureka.co/big-data-and-hadoop
Read 1 TB Data
4 I/O Channels
Each Channel – 100 MB/s
1 Machine
4 I/O Channels
Each Channel – 100 MB/s
10 Machine
Slide 20Slide 20Slide 20 www.edureka.co/big-data-and-hadoop
4 I/O Channels
Each Channel – 100 MB/s
1 Machine
4 I/O Channels
Each Channel – 100 MB/s
10 Machine
43 Minutes
Read 1 TB Data
Slide 21Slide 21Slide 21 www.edureka.co/big-data-and-hadoop
4 I/O Channels
Each Channel – 100 MB/s
1 Machine
4 I/O Channels
Each Channel – 100 MB/s
10 Machine
4.3 Minutes43 Minutes
Read 1 TB Data
Slide 22Slide 22Slide 22 www.edureka.co/big-data-and-hadoop
→ Apache Hadoop is a framework that allows for the distributed processing of large data sets across clusters
of commodity computers using a simple programming model.
→ It is an Open-source Data Management with scale-out storage and distributed processing.
Slide 23 www.edureka.co/big-data-and-hadoop
Hadoop is a framework that allows for the distributed
processing of:
» Small Data Sets
» Large Data Sets
Slide 24 www.edureka.co/big-data-and-hadoop
Ans. Large Data Sets.
It is also capable of processing small data-sets. However, to
experience the true power of Hadoop, one needs to have
data in TB’s. Because this is where RDBMS takes hours and
fails whereas Hadoop does the same in couple of minutes.
Slide 25Slide 25Slide 25 www.edureka.co/big-data-and-hadoop
Pig Latin
Data Analysis
Hive
DW System
Other YARN
Frameworks
(MPI, GRAPH)
HBaseMapReduce Framework
YARN
Cluster Resource Management
Apache Oozie
(Workflow)
HDFS
(Hadoop Distributed File System)
Hadoop 2.0
Sqoop
Unstructured or
Semi-structured Data Structured Data
Flume
Mahout
Machine Learning
Slide 26Slide 26Slide 26 www.edureka.co/big-data-and-hadoop
DataNode
Node
Manager
DataNode DataNode DataNode
YARN
HDFS
Cluster
Resource
Manager
NameNode
Node
Manager
Node
Manager
Node
Manager
Slide 27 www.edureka.co/big-data-and-hadoop
RAM: 16GB
Hard disk: 6 x 2TB
Processor: Xenon with 2 cores
Ethernet: 3 x 10 GB/s
OS: 64-bit CentOS
RAM: 16GB
Hard disk: 6 x 2TB
Processor: Xenon with 2 cores.
Ethernet: 3 x 10 GB/s
OS: 64-bit CentOS
RAM: 64 GB,
Hard disk: 1 TB
Processor: Xenon with 8 Cores
Ethernet: 3 x 10 GB/s
OS: 64-bit CentOS
Power: Redundant Power Supply
RAM: 32 GB,
Hard disk: 1 TB
Processor: Xenon with 4 Cores
Ethernet: 3 x 10 GB/s
OS: 64-bit CentOS
Power: Redundant Power Supply
Active NameNodeSecondary NameNode
DataNode DataNode
RAM: 64 GB,
Hard disk: 1 TB
Processor: Xenon with 8 Cores
Ethernet: 3 x 10 GB/s
OS: 64-bit CentOS
Power: Redundant Power Supply
StandBy NameNode
Slide 28 www.edureka.co/big-data-and-hadoop
Master
NameNode
http://master:50070/
ResourceManager
http://master:8088
Slave01
DataNode
NodeManager
Slave02
DataNode
NodeManager
Slave03
DataNode
NodeManager
Slave04
DataNode
NodeManager
Slave05
DataNode
NodeManager
Slide 29 www.edureka.co/big-data-and-hadoop
NodeManager
DataNode
NodeManager
HDFS YARN
NameNode
DataNode
NodeManager DataNode
ResourceManager
DataNode
NodeManager
DataNode
NodeManager
NodeManager
DataNode
NodeManager
DataNode
NodeManager
DataNode
Slide 30 www.edureka.co/big-data-and-hadoop
Namenode NS
Storage
…
NamespaceBlockStorage
Namespace
NN-1 NN-k NN-n
Common Storage
BlockStorage
Pool 1 Pool k Pool n
Block Pools
… …
Hadoop 1.0 Hadoop 2.0
DatanodeDatanode
Datanode 1
…
Datanode m
…
Datanode 2
…
Block Management
Slide 31 www.edureka.co/big-data-and-hadoop
How does HDFS Federation help HDFS Scale horizontally?
a. Reduces the load on any single NameNode by using the multiple,
independent NameNode to manage individual parts of the file system
namespace.
b. Provides cross-data centre (non-local) support for HDFS, allowing a
cluster administrator to split the Block Storage outside the local cluster.
Slide 32 www.edureka.co/big-data-and-hadoop
Ans. Option (a)
In order to scale the name service horizontally, HDFS federation
uses multiple independent NameNode. The NameNode are
federated, that is, the NameNode are independent and don’t
require coordination with each other.
Slide 33 www.edureka.co/big-data-and-hadoop
You have configured two name nodes to manage /marketing and
/finance respectively. What will happen if you try to put a file in
/accounting directory?
Slide 34 www.edureka.co/big-data-and-hadoop
Ans. Put will fail. None of the namespace will manage the file and you
will get an IOException with a No such file or directory error.
Slide 35 www.edureka.co/big-data-and-hadoop
Node Manager
Container
App
Master
Node Manager
Container
App
Master
HDFS YARN
Resource
Manager
All name space edits
logged to shared NFS
storage; single writer
(fencing)
Read edit logs and
applies to its own
namespace
Secondary
Name Node
DataNode
Standby
NameNode
Active
NameNode
DataNode Data Node
DataNodeDataNode
NameNode
High
Availability
Next Generation
MapReduce
*Not necessary to
configure
Secondary
NameNode
Client
Shared Edit Logs
HDFS HIGH AVAILABILITY
Node Manager
Container
App
Master
Node Manager
Container
App
Master
Slide 36 www.edureka.co/big-data-and-hadoop
Node Manager
Container
App
Master
Node Manager
Container
App
Master
HDFS YARN
Resource
Manager
All name space edits
logged to shared NFS
storage; single writer
(fencing)
Read edit logs and
applies to its own
namespace
Secondary
Name Node
DataNode
Standby
NameNode
Active
NameNode
DataNode Data Node
DataNodeDataNode
NameNode
High
Availability
Next Generation
MapReduce
*Not necessary to
configure
Secondary
NameNode
Client
Shared Edit Logs
HDFS HIGH AVAILABILITY
Node Manager
Container
App
Master
Node Manager
Container
App
Master
Slide 37 www.edureka.co/big-data-and-hadoop
HDFS HA was developed to overcome the following disadvantage in Hadoop
1.0?
a. Single Point of Failure of Name-Node
b. Only one version can be run in classic Map-Reduce
c. Too much burden on Job Tracker
Slide 38 www.edureka.co/big-data-and-hadoop
Ans. Single Point of Failure of NameNode
Slide 39 www.edureka.co/big-data-and-hadoop
…
Slide 40 www.edureka.co/big-data-and-hadoop
Facebook
→ We use Hadoop to store copies of internal log and dimension data sources and use
it as a source for reporting/analytics and machine learning.
→ Currently we have 2 major clusters:
» A 1100-machine cluster with 8800 cores and about 12 PB raw storage.
» A 300-machine cluster with 2400 cores and about 3 PB raw storage.
» Each (commodity) node has 8 cores and 12 TB of storage.
» We are heavy users of both streaming as well as the Java APIs. We have built
a higher level data warehousing framework using these features called Hive
(see the http://Hadoop.apache.org/hive/). We have also developed a FUSE
implementation over HDFS.
Slide 41 www.edureka.co/big-data-and-hadoop
Hadoop can run in any of the following three modes:
Fully-Distributed Mode
Pseudo-Distributed Mode
→ No daemons, everything runs in a single JVM.
→ Suitable for running MapReduce programs during development.
→ Has no DFS.
→ Hadoop daemons run on the local machine.
→ Hadoop daemons run on a cluster of machines.
Standalone (or Local) Mode
Slide 42 www.edureka.co/big-data-and-hadoop
→ Apache Hadoop and HDFS
→ http://www.edureka.in/blog/introduction-to-apache-hadoop-hdfs/
→ Apache Hadoop HDFS Architecture
→ http://www.edureka.in/blog/apache-hadoop-hdfs-architecture/
Slide 43 www.edureka.co/big-data-and-hadoop
• Referring the documents present in the LMS under assignment solve the below problem.
How many such DataNodes you would need to read 100TB data in 5 minutes in your Hadoop Cluster?
Slide 44
Your feedback is important to us, be it a compliment, a suggestion or a complaint. It helps us to make
the course better!
Please spare few minutes to take the survey after the webinar.
www.edureka.co/big-data-and-hadoop
Introduction to Big Data & Hadoop

Introduction to Big Data & Hadoop

  • 1.
  • 2.
  • 3.
    Slide 3 www.edureka.co/big-data-and-hadoop Atthe end of this session , you will understand the: → Big Data Learning Paths → Big Data Introduction → Hadoop and Its Eco-System → Hadoop Architecture → Next Step on How to Setup Hadoop
  • 4.
    Slide 4 www.edureka.co/big-data-and-hadoop •Java / Python / Ruby • Hadoop Eco-system • NoSQL DB • Spark • Linux Administration • Cluster Management • Cluster Performance • Virtualization • Statistics Skills • Machine Learning • Hadoop Essentials • Expertise in R Developer/Testing Administration Data Analyst Big Data and Hadoop MapReduce Design Patterns Apache Spark & Scala Apache Cassandra Linux Administration Hadoop Administration Data Science Business Analytics Using R Advance Predictive Modelling in R Talend for Big Data Data Visualization Using Tableau
  • 5.
    Slide 5 www.edureka.co/big-data-and-hadoop →Lots of Data (Terabytes or Petabytes) → Big data is the term for a collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools or traditional data processing applications → The challenges include capture, curation, storage, search, sharing, transfer, analysis, and visualization Big Data
  • 6.
    Slide 6 www.edureka.co/big-data-and-hadoop →Systems / Enterprises generate huge amount of data from Terabytes to Petabytes of information Stock market generates about one terabyte of new trade data per day to perform stock trading analytics to determine trends for optimal trades
  • 7.
    Slide 7 www.edureka.co/big-data-and-hadoop →By 2020, IDC (International Data Corporation) predicts the number will have reached 40,000 EB, or 40 Zettabytes (ZB) → The world’s information is doubling every two years. By 2020 the world will generate 50 times the amount of information and 75 times the number of information containers
  • 8.
    Slide 8 www.edureka.co/big-data-and-hadoop IBM’sDefinition – Big Data Characteristics http://www-01.ibm.com/software/data/bigdata/ VOLUME Web logs Images Videos Audios Sensor Data VARIET Y VELOCITY VERACITY Min Max Mean SD 4.3 7.9 5.84 0.83 2.0 4.4 3.05 0.43 0.1 2.5 1.20 0.76
  • 9.
    Slide 9 www.edureka.co/big-data-and-hadoop HelloThere!! My name is Annie. I love quizzes and puzzles and I am here to make you guys think and answer my questions.
  • 10.
    Slide 10 www.edureka.co/big-data-and-hadoop Mapthe following to corresponding data type: » XML files, e-mail body » Audio, Video, Images, Archived documents » Data from Enterprise systems (ERP, CRM etc.)
  • 11.
    Slide 11 www.edureka.co/big-data-and-hadoop Ans.XML files, e-mail body → Semi-structured data Audio, Video, Image, Files, Archived documents → Unstructured data Data from Enterprise systems (ERP, CRM etc.) → Structured data
  • 12.
    Slide 12 www.edureka.co/big-data-and-hadoop Moreon Big Data • http://www.edureka.in/blog/the-hype-behind-big-data/ Why Hadoop? • http://www.edureka.in/blog/why-hadoop/ Opportunities in Hadoop • http://www.edureka.in/blog/jobs-in-hadoop/ Big Data • http://en.wikipedia.org/wiki/Big_Data IBM’s definition – Big Data Characteristics • http://www-01.ibm.com/software/data/bigdata/
  • 13.
    Slide 13Slide 13Slide13 www.edureka.co/big-data-and-hadoop → Web and e-tailing » Recommendation Engines » Ad Targeting » Search Quality » Abuse and Click Fraud Detection → Telecommunications » Customer Churn Prevention » Network Performance Optimization » Calling Data Record (CDR) Analysis » Analysing Network to Predict Failure http://wiki.apache.org/hadoop/PoweredBy
  • 14.
    Slide 14Slide 14Slide14 www.edureka.co/big-data-and-hadoop → Government » Fraud Detection and Cyber Security » Welfare Schemes » Justice → Healthcare and Life Sciences » Health Information Exchange » Gene Sequencing » Serialization » Healthcare Service Quality Improvements » Drug Safety http://wiki.apache.org/hadoop/PoweredBy
  • 15.
    Slide 15Slide 15Slide15 www.edureka.co/big-data-and-hadoop → Banks and Financial services » Modeling True Risk » Threat Analysis » Fraud Detection » Trade Surveillance » Credit Scoring and Analysis → Retail » Point of Sales Transaction Analysis » Customer Churn Analysis » Sentiment Analysis http://wiki.apache.org/hadoop/PoweredBy
  • 16.
    Slide 16Slide 16Slide16 www.edureka.co/big-data-and-hadoop → Insight into data can provide Business Advantage. → Some key early indicators can mean Fortunes to Business. → More Precise Analysis with more data. *Sears was using traditional systems such as Oracle Exadata, Teradata and SAS etc., to store and process the customer activity and sales data. Case Study: Sears Holding Corporation
  • 17.
    Slide 17Slide 17Slide17 www.edureka.co/big-data-and-hadoop Mostly Append BI Reports + Interactive Apps RDBMS (Aggregated Data) ETL Compute Grid Storage only Grid (Original Raw Data) Collection Inctrumentation A meagre 10% of the ~2PB data is available for BI Storage 2. Moving data to compute doesn’ t scale 90% of the ~2PB archived Processing 3. Premature data death 1. Can’t explore original high fidelity raw data
  • 18.
    Slide 18Slide 18Slide18 www.edureka.co/big-data-and-hadoop Mostly Append BI Reports + Interactive Apps RDBMS (Aggregated Data) Hadoop : Storage + Compute Grid Collection Instrumentation Both Storage And Processing Entire ~2PB Data is available for processing No Data Archiving 1. Data Exploration & Advanced analytics 2. Scalable throughput for ETL & aggregation 3. Keep data alive forever *Sears moved to a 300-Node Hadoop cluster to keep 100% of its data available for processing rather than a meagre 10% as was the case with existing Non-Hadoop solutions.
  • 19.
    Slide 19Slide 19Slide19 www.edureka.co/big-data-and-hadoop Read 1 TB Data 4 I/O Channels Each Channel – 100 MB/s 1 Machine 4 I/O Channels Each Channel – 100 MB/s 10 Machine
  • 20.
    Slide 20Slide 20Slide20 www.edureka.co/big-data-and-hadoop 4 I/O Channels Each Channel – 100 MB/s 1 Machine 4 I/O Channels Each Channel – 100 MB/s 10 Machine 43 Minutes Read 1 TB Data
  • 21.
    Slide 21Slide 21Slide21 www.edureka.co/big-data-and-hadoop 4 I/O Channels Each Channel – 100 MB/s 1 Machine 4 I/O Channels Each Channel – 100 MB/s 10 Machine 4.3 Minutes43 Minutes Read 1 TB Data
  • 22.
    Slide 22Slide 22Slide22 www.edureka.co/big-data-and-hadoop → Apache Hadoop is a framework that allows for the distributed processing of large data sets across clusters of commodity computers using a simple programming model. → It is an Open-source Data Management with scale-out storage and distributed processing.
  • 23.
    Slide 23 www.edureka.co/big-data-and-hadoop Hadoopis a framework that allows for the distributed processing of: » Small Data Sets » Large Data Sets
  • 24.
    Slide 24 www.edureka.co/big-data-and-hadoop Ans.Large Data Sets. It is also capable of processing small data-sets. However, to experience the true power of Hadoop, one needs to have data in TB’s. Because this is where RDBMS takes hours and fails whereas Hadoop does the same in couple of minutes.
  • 25.
    Slide 25Slide 25Slide25 www.edureka.co/big-data-and-hadoop Pig Latin Data Analysis Hive DW System Other YARN Frameworks (MPI, GRAPH) HBaseMapReduce Framework YARN Cluster Resource Management Apache Oozie (Workflow) HDFS (Hadoop Distributed File System) Hadoop 2.0 Sqoop Unstructured or Semi-structured Data Structured Data Flume Mahout Machine Learning
  • 26.
    Slide 26Slide 26Slide26 www.edureka.co/big-data-and-hadoop DataNode Node Manager DataNode DataNode DataNode YARN HDFS Cluster Resource Manager NameNode Node Manager Node Manager Node Manager
  • 27.
    Slide 27 www.edureka.co/big-data-and-hadoop RAM:16GB Hard disk: 6 x 2TB Processor: Xenon with 2 cores Ethernet: 3 x 10 GB/s OS: 64-bit CentOS RAM: 16GB Hard disk: 6 x 2TB Processor: Xenon with 2 cores. Ethernet: 3 x 10 GB/s OS: 64-bit CentOS RAM: 64 GB, Hard disk: 1 TB Processor: Xenon with 8 Cores Ethernet: 3 x 10 GB/s OS: 64-bit CentOS Power: Redundant Power Supply RAM: 32 GB, Hard disk: 1 TB Processor: Xenon with 4 Cores Ethernet: 3 x 10 GB/s OS: 64-bit CentOS Power: Redundant Power Supply Active NameNodeSecondary NameNode DataNode DataNode RAM: 64 GB, Hard disk: 1 TB Processor: Xenon with 8 Cores Ethernet: 3 x 10 GB/s OS: 64-bit CentOS Power: Redundant Power Supply StandBy NameNode
  • 28.
  • 29.
    Slide 29 www.edureka.co/big-data-and-hadoop NodeManager DataNode NodeManager HDFSYARN NameNode DataNode NodeManager DataNode ResourceManager DataNode NodeManager DataNode NodeManager NodeManager DataNode NodeManager DataNode NodeManager DataNode
  • 30.
    Slide 30 www.edureka.co/big-data-and-hadoop NamenodeNS Storage … NamespaceBlockStorage Namespace NN-1 NN-k NN-n Common Storage BlockStorage Pool 1 Pool k Pool n Block Pools … … Hadoop 1.0 Hadoop 2.0 DatanodeDatanode Datanode 1 … Datanode m … Datanode 2 … Block Management
  • 31.
    Slide 31 www.edureka.co/big-data-and-hadoop Howdoes HDFS Federation help HDFS Scale horizontally? a. Reduces the load on any single NameNode by using the multiple, independent NameNode to manage individual parts of the file system namespace. b. Provides cross-data centre (non-local) support for HDFS, allowing a cluster administrator to split the Block Storage outside the local cluster.
  • 32.
    Slide 32 www.edureka.co/big-data-and-hadoop Ans.Option (a) In order to scale the name service horizontally, HDFS federation uses multiple independent NameNode. The NameNode are federated, that is, the NameNode are independent and don’t require coordination with each other.
  • 33.
    Slide 33 www.edureka.co/big-data-and-hadoop Youhave configured two name nodes to manage /marketing and /finance respectively. What will happen if you try to put a file in /accounting directory?
  • 34.
    Slide 34 www.edureka.co/big-data-and-hadoop Ans.Put will fail. None of the namespace will manage the file and you will get an IOException with a No such file or directory error.
  • 35.
    Slide 35 www.edureka.co/big-data-and-hadoop NodeManager Container App Master Node Manager Container App Master HDFS YARN Resource Manager All name space edits logged to shared NFS storage; single writer (fencing) Read edit logs and applies to its own namespace Secondary Name Node DataNode Standby NameNode Active NameNode DataNode Data Node DataNodeDataNode NameNode High Availability Next Generation MapReduce *Not necessary to configure Secondary NameNode Client Shared Edit Logs HDFS HIGH AVAILABILITY Node Manager Container App Master Node Manager Container App Master
  • 36.
    Slide 36 www.edureka.co/big-data-and-hadoop NodeManager Container App Master Node Manager Container App Master HDFS YARN Resource Manager All name space edits logged to shared NFS storage; single writer (fencing) Read edit logs and applies to its own namespace Secondary Name Node DataNode Standby NameNode Active NameNode DataNode Data Node DataNodeDataNode NameNode High Availability Next Generation MapReduce *Not necessary to configure Secondary NameNode Client Shared Edit Logs HDFS HIGH AVAILABILITY Node Manager Container App Master Node Manager Container App Master
  • 37.
    Slide 37 www.edureka.co/big-data-and-hadoop HDFSHA was developed to overcome the following disadvantage in Hadoop 1.0? a. Single Point of Failure of Name-Node b. Only one version can be run in classic Map-Reduce c. Too much burden on Job Tracker
  • 38.
    Slide 38 www.edureka.co/big-data-and-hadoop Ans.Single Point of Failure of NameNode
  • 39.
  • 40.
    Slide 40 www.edureka.co/big-data-and-hadoop Facebook →We use Hadoop to store copies of internal log and dimension data sources and use it as a source for reporting/analytics and machine learning. → Currently we have 2 major clusters: » A 1100-machine cluster with 8800 cores and about 12 PB raw storage. » A 300-machine cluster with 2400 cores and about 3 PB raw storage. » Each (commodity) node has 8 cores and 12 TB of storage. » We are heavy users of both streaming as well as the Java APIs. We have built a higher level data warehousing framework using these features called Hive (see the http://Hadoop.apache.org/hive/). We have also developed a FUSE implementation over HDFS.
  • 41.
    Slide 41 www.edureka.co/big-data-and-hadoop Hadoopcan run in any of the following three modes: Fully-Distributed Mode Pseudo-Distributed Mode → No daemons, everything runs in a single JVM. → Suitable for running MapReduce programs during development. → Has no DFS. → Hadoop daemons run on the local machine. → Hadoop daemons run on a cluster of machines. Standalone (or Local) Mode
  • 42.
    Slide 42 www.edureka.co/big-data-and-hadoop →Apache Hadoop and HDFS → http://www.edureka.in/blog/introduction-to-apache-hadoop-hdfs/ → Apache Hadoop HDFS Architecture → http://www.edureka.in/blog/apache-hadoop-hdfs-architecture/
  • 43.
    Slide 43 www.edureka.co/big-data-and-hadoop •Referring the documents present in the LMS under assignment solve the below problem. How many such DataNodes you would need to read 100TB data in 5 minutes in your Hadoop Cluster?
  • 44.
    Slide 44 Your feedbackis important to us, be it a compliment, a suggestion or a complaint. It helps us to make the course better! Please spare few minutes to take the survey after the webinar. www.edureka.co/big-data-and-hadoop