1. Presented by,
Atul Kushwaha
B.Tech, 3rd
Year(IT)
1109113033

2. Introduction
Big Data:
•Big data is a term used to describe the voluminous
amount of unstructured and semi-structured data a
company creates.
•Data that would take too much time and cost too much
money to load into a relational database for analysis.
• Big data doesn't refer to any specific quantity, the term
is often used when speaking about petabytes and
exabytes of data.

3. • The New York Stock Exchange generates about one terabyte of new
trade data per day.
• Facebook hosts approximately 10 billion photos, taking up one
petabyte of storage.
• The Internet Archive stores around 2 petabytes of data, and is
growing at a rate of 20 terabytes per month.
• The Large Hadron Collider near Geneva, Switzerland, produces about
15 petabytes of data per year.
Facts!!

4. So What Is The Problem?
 The transfer speed is around 100 MB/s
 A standard disk is 1 Terabyte
 Time to read entire disk= 10000 seconds or 3 Hours!
 Increase in processing time may not be as helpful because
• Network bandwidth is now more of a limiting factor
• Physical limits of processor chips have been reached

5. So What do We Do?
•The obvious solution is that
we use multiple processors to
solve the same problem by
fragmenting it into pieces.
•Imagine if we had 100 drives,
each holding one hundredth
of the data. Working in
parallel, we could read the
data in under two minutes.

6. Distributed Computing Vs
Parallelization
 Parallelization- Multiple processors or CPU’s
in a single machine
 Distributed Computing- Multiple computers
connected via a network

7. Problems In Distributed Computing
• Hardware Failure:
As soon as we start using many pieces of
hardware, the chance that one will fail is fairly
high.
• Combine the data after analysis:
Most analysis tasks need to be able to combine
the data in some way; data read from one
disk may need to be combined with the data
from any of the other 99 disks.

8. To The Rescue!
Apache Hadoop is a framework for running applications on large
cluster built of commodity hardware.
A common way of avoiding data loss is through replication:
redundant copies of the data are kept by the system so that in the
event of failure, there is another copy available. The Hadoop
Distributed Filesystem (HDFS), takes care of this problem.
The second problem is solved by a simple programming model-
Mapreduce. Hadoop is the popular open source implementation
of MapReduce, a powerful tool designed for deep analysis and
transformation of very large data sets.

9. What is ? ?
• It is an open source project by the Apache Foundation to handle
large data processing
• It was inspired by Google’s MapReduce and Google File System
(GFS) papers in 2003 and 2004
• It was originally conceived by Doug Cutting in 2005 and first used by
Yahoo! in 2006
• It is named after his son’s pet elephant incidentally
• It is basically a distributed file system which is written in Java.

10. Hadoop Approach to Distributed
Computing
 The theoretical 1000-CPU machine would cost a very
large amount of money, far more than 1,000 single-CPU.
 Hadoop will tie these smaller and more reasonably priced
machines together into a single cost-effective computer
cluster.
 Hadoop provides a simplified programming model which
allows the user to quickly write and test distributed
systems, and its’ efficient, automatic distribution of data
and work across machines and in turn utilizing the
underlying parallelism of the CPU cores.

11. Hadoop Components
HDFSHDFS
Storage
Self-healing
high-bandwidth
clustered storage
MapReduceMapReduce
Processing
Fault-tolerant
distributed
processing

12. Hadoop MapReduce
 MapReduce is a programming model
 Programs written in this functional style are automatically parallelized and
executed on a large cluster of commodity machines
 MapReduce is an associated implementation for processing and generating
large data sets.

13. The Programming Model Of MapReduce
 Map, written by the user, takes an input pair and produces a set of intermediate
key/value pairs. The MapReduce library groups together all intermediate values
associated with the same intermediate key I and passes them to the Reduce
function.

14.  The Reduce function, also written by the user, accepts an intermediate key I and a set of values
for that key. It merges together these values to form a possibly smaller set of values

15.  Filesystems that manage the storage across a network of machines
are called distributed filesystems.
 Hadoop comes with a distributed filesystem called HDFS, which
stands for Hadoop Distributed Filesystem.
 HDFS, the Hadoop Distributed File System, is a distributed file
system designed to hold very large amounts of data (terabytes or even
petabytes), and provide high-throughput access to this information.
 The Hadoop distributed file system (HDFS) is a distributed,
scalable, and portable file-system written in Java for the Hadoop
framework.
HADOOP DISTRIBUTED
FILESYSTEM (HDFS)

16. HDFS
 It manages storage on the cluster by breaking
incoming files into pieces, called blocks
 Stores each of the blocks redundantly across the
pool of servers
 It stores three complete copies of each file by
copying each piece to three different servers

17. Namenodes and Datanodes
 A HDFS cluster has two types of node operating in a master-slave
pattern: a namenode (the master) and a number of datanodes (slave).
 The namenode manages the filesystem namespace. It maintains the
filesystem tree and the metadata for all the files and directories in the
tree.
 Datanodes are the work horses of the filesystem.It manages storage
attached to the nodes that they run on.
 HDFS exposes a file system namespace and allows user data to be
stored in files.
 Internally, a file is split into one or more blocks and these blocks are
stored in a set of DataNodes.

19. Who Uses Hadoop?

20. Advantages Over RDBMS
• Scalable: It can reliably store and process petabytes.
• Economical: It distributes the data and processing across
clusters of commonly available computers (in
thousands).
• Efficient: By distributing the data, it can process it in
parallel on the nodes where the data is located.
• Reliable: It automatically maintains multiple copies of
data and automatically redeploys computing tasks based
on failures.

21. Conclusion
 So major companies like facebook
amazon,yahoo,etc. are adapting Hadoop and in
future there can be many names in the list.
 This technology has bright future scope because day
by day need of data would increase and security
issues also the major point.
 Hence Hadoop Technology is the best appropriate
approach for handling the large data in smart way
and its future is bright…

22. Questions???
‘Thank You’