Learn what is Hadoop-and-BigData

Hadoop Video/Online Training by Expert
Contact Us:
India: 8121660044
USA : 732-419-2619
Site: http://www.hadooponlinetutor.com

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.
http://www.hadooponlinetutor.com

• 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.
• Ancestry.com, the genealogy site, stores around 2.5 petabytes of data.
• 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.

What Caused The Problem?
1
2
Year
Standard Hard Drive Size
(in Mb)
1990 1370
2010 1000000
1
2
Year
Data Transfer Rate
(Mbps)
1990 4.4
2010 100

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

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.

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

Examples
Cray-2 was a four-processor ECL
vector supercomputer made by
Cray Research starting in 1985

Distributed Computing
The key issues involved in this Solution:
 Hardware failure
 Combine the data after analysis
 Network Associated Problems

What Can We Do With A Distributed
Computer System?
 IBM Deep Blue
 Multiplying Large Matrices
 Simulating several 100’s of characters-
LOTRs
 Index the Web (Google)
 Simulating an internet size network for
network experiments

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.

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.http://www.hadooponlinetutor.com

What Else is Hadoop?
A reliable shared storage and analysis system.
There are other subprojects of Hadoop that provide complementary
services, or build on the core to add higher-level abstractions The various
subprojects of hadoop include:
1. Core
2. Avro
3. Pig
4. HBase
5. Zookeeper
6. Hive
7. Chukwa

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 compute 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.

MapReduce

MapReduce
 Hadoop limits the amount of communication which can be performed by the
processes, as each individual record is processed by a task in isolation from one another
 By restricting the communication between nodes, Hadoop makes the distributed system
much more reliable. Individual node failures can be worked around by restarting tasks
on other machines.
 The other workers continue to operate as though nothing went wrong, leaving the
challenging aspects of partially restarting the program to the underlying Hadoop layer.
Map : (in_value,in_key)(out_key, intermediate_value)
Reduce: (out_key, intermediate_value) (out_value list)

What is 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.
MapReduce
MAP
map function that
processes a key/value
pair to generate a set of
intermediate key/value
pairs
REDUCE
and a reduce function
that merges all
intermediate values
associated with the same
intermediate key.

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.

 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

 This abstraction allows us to handle lists of values that are too large to fit in memory.
 Example:
 // key: document name
 // value: document contents
 for each word w in value:
 EmitIntermediate(w, "1");
 reduce(String key, Iterator values):
 // key: a word
 // values: a list of counts
 int result = 0;
 for each v in values:
 result += ParseInt(v);
 Emit(AsString(result));

Orientation of Nodes
Data Locality Optimization:
The computer nodes and the storage nodes are the same. The Map-Reduce
framework and the Distributed File System run on the same set of nodes. This
configuration allows the framework to effectively schedule tasks on the nodes where
data is already present, resulting in very high aggregate bandwidth across the
cluster.
If this is not possible: The computation is done by another processor on the same
rack.
“Moving Computation is Cheaper than Moving Data”

How MapReduce Works
 A Map-Reduce job usually splits the input data-set into independent chunks which are
processed by the map tasks in a completely parallel manner.
 The framework sorts the outputs of the maps, which are then input to the reduce tasks.
 Typically both the input and the output of the job are stored in a file-system. The
framework takes care of scheduling tasks, monitoring them and re-executes the failed
tasks.
 A MapReduce job is a unit of work that the client wants to be performed: it consists of
the input data, the MapReduce program, and configuration information. Hadoop runs
the job by dividing it into tasks, of which there are two types: map tasks and reduce
tasks

Fault Tolerance
 There are two types of nodes that control the job execution process: tasktrackers and
jobtrackers
 The jobtracker coordinates all the jobs run on the system by scheduling tasks to run on
tasktrackers.
 Tasktrackers run tasks and send progress reports to the jobtracker, which keeps a record
of the overall progress of each job.
 If a tasks fails, the jobtracker can reschedule it on a different tasktracker.

Input Splits
 Input splits: Hadoop divides the input to a MapReduce job into fixed-size
pieces called input splits, or just splits. Hadoop creates one map task for each
split, which runs the user-defined map function for each record in the split.
 The quality of the load balancing increases as the splits become more fine-
grained.
 BUT if splits are too small, then the overhead of managing the splits and of map
task creation begins to dominate the total job execution time. For most jobs, a
good split size tends to be the size of a HDFS block, 64 MB by default.
WHY?
 Map tasks write their output to local disk, not to HDFS. Map output is
intermediate output: it’s processed by reduce tasks to produce the final output,
and once the job is complete the map output can be thrown away. So storing it
in HDFS, with replication, would be a waste of time. It is also possible that the
node running the map task fails before the map output has been consumed by
the reduce task.

Input to Reduce Tasks
 Reduce tasks don’t have the advantage of
data locality—the input to a single reduce
task is normally the output from all mappers.

MapReduce data flow with a single reduce task

MapReduce data flow with multiple reduce tasks

MapReduce data flow with no reduce tasks

•Many MapReduce jobs are limited by the bandwidth available on the cluster.
•In order to minimize the data transferred between the map and reduce tasks, combiner
functions are introduced.
•Hadoop allows the user to specify a combiner function to be run on the map output—the
combiner function’s output forms the input to the reduce function.
•Combiner finctions can help cut down the amount of data shuffled between the maps and
the reduces.
Combiner Functions

•Hadoop provides an API to MapReduce that allows you to
write your map and reduce functions in languages other
than Java.
•Hadoop Streaming uses Unix standard streams as the
interface between Hadoop and your program, so you can use
any language that can read standard input and write to
standard output to write your MapReduce program.
Hadoop Streaming:

•Hadoop Pipes is the name of the C++ interface to Hadoop MapReduce.
•Unlike Streaming, which uses standard input and output to communicate with
the map and reduce code, Pipes uses sockets as the channel over which the
tasktracker communicates with the process running the C++ map or reduce
function. JNI is not used.
Hadoop Pipes:

 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.
HADOOP DISTRIBUTED
FILESYSTEM (HDFS)

Problems In Distributed File Systems
Making distributed filesystems is more complex than regular disk filesystems. This
is because the data is spanned over multiple nodes, so all the complications of
network programming kick in.
•Hardware Failure
An HDFS instance may consist of hundreds or thousands of server machines, each storing
part of the file system’s data. The fact that there are a huge number of components and that
each component has a non-trivial probability of failure means that some component of HDFS
is always non-functional. Therefore, detection of faults and quick, automatic recovery from
them is a core architectural goal of HDFS.
•Large Data Sets
Applications that run on HDFS have large data sets. A typical file in HDFS is gigabytes to
terabytes in size. Thus, HDFS is tuned to support large files. It should provide high
aggregate data bandwidth and scale to hundreds of nodes in a single cluster. It should
support tens of millions of files in a single instance.

Goals of HDFS
Streaming Data Access
Applications that run on HDFS need streaming access to their data sets. They are
not general purpose applications that typically run on general purpose file systems.
HDFS is designed more for batch processing rather than interactive use by users.
The emphasis is on high throughput of data access rather than low latency of data
access. POSIX imposes many hard requirements that are not needed for
applications that are targeted for HDFS. POSIX semantics in a few key areas has
been traded to increase data throughput rates.
Simple Coherency Model
HDFS applications need a write-once-read-many access model for files. A file once
created, written, and closed need not be changed. This assumption simplifies data
coherency issues and enables high throughput data access. A Map/Reduce
application or a web crawler application fits perfectly with this model. There is a plan
to support appending-writes to files in the future.

“Moving Computation is Cheaper than Moving Data”
 A computation requested by an application is much more efficient if
it is executed near the data it operates on. This is especially true when
the size of the data set is huge. This minimizes network congestion
and increases the overall throughput of the system. The assumption is
that it is often better to migrate the computation closer to where the
data is located rather than moving the data to where the application is
running. HDFS provides interfaces for applications to move
themselves closer to where the data is located.
Portability Across Heterogeneous Hardware and Software
Platforms HDFS has been designed to be easily portable from
one platform to another. This facilitates widespread adoption
of HDFS as a platform of choice for a large set of
applications.

Design of HDFS
 Very large files
Files that are hundreds of megabytes, gigabytes, or terabytes in size. There
are Hadoop clusters running today that store petabytes of data.
 Streaming data access
HDFS is built around the idea that the most efficient data processing pattern
is a write-once, read-many-times pattern.
A dataset is typically generated or copied from source, then various
analyses are performed on that dataset over time. Each analysis will involve
a large proportion of the dataset, so the time to read the whole dataset is
more important than the latency in reading the first record.

 Low-latency data access
Applications that require low-latency access to data, in the tens
of milliseconds
range, will not work well with HDFS. Remember HDFS is
optimized for delivering a high throughput of data, and this may
be at the expense of latency. HBase (Chapter 12) is currently a
better choice for low-latency access.
 Multiple writers, arbitrary file modifications
Files in HDFS may be written to by a single writer. Writes are
always made at the end of the file. There is no support for
multiple writers, or for modifications at arbitrary offsets in the
file. (These might be supported in the future, but they are likely
to be relatively inefficient.)

• Lots of small files
Since the namenode holds filesystem metadata in memory, the limit to
the number of files in a filesystem is governed by the amount of
memory on the namenode. As a rule of thumb, each file, directory, and
block takes about 150 bytes. So, for example, if you had one million
files, each taking one block, you would need at least 300 MB of
memory. While storing millions of files is feasible, billions is beyond the
capability of current hardware.

 Commodity hardware
Hadoop doesn’t require expensive, highly reliable hardware to run on.
It’s designed to run on clusters of commodity hardware for which the
chance of node failure across the cluster is high, at least for large
clusters. HDFS is designed to carry on working without a noticeable
interruption to the user in the face of such failure. It is also worth
examining the applications for which using HDFS does not work so
well. While this may change in the future, these are areas where
HDFS is not a good fit today:

Contact Us:
Our Address:
#444, 4th floor, Gumidelli Commercial Complex
Reliance Trends Building
Begumpet, Hyderabad
Phone:
USA : +1 732-419-2619
INDIA: +91 8121660044
Email:
srini.itlm@gmail.com
Website: http://www.hadooponlinetutor.com

Learn what is Hadoop-and-BigData

More Related Content

What's hot (20)

Similar to Learn what is Hadoop-and-BigData (20)

Recently uploaded (20)

Learn what is Hadoop-and-BigData