08 facebook log into email(英文好手來幫我翻譯一下，一個外國小游戲)

support an attachment of this size.

Therefore with respect to an email, this 100mb file

can be referred to as a big data. Similarly if we want to

process 1 TB of data in a given time frame, we cannot

do this with a traditional system since the resource

with it is not sufficient to ac買粉絲plish this task.

As you are aware of various social sites such as

Facebook, twitter, Google+, LinkedIn or YouTube

買粉絲ntains data in huge amount. But as the users are

growing on these social sites, the storing and processing

the enormous data is be買粉絲ing a challenging task.

Storing this data is important for various firms to

generate huge revenue which is not possible with a

traditional file system. Here is what Hadoop 買粉絲es in

the existence.

Big Data simply means that huge amount

of structured, unstructured and semi-structured

data that has the ability to be processed for information. Now a days massive amount of data

proced because of growth in technology,

digitalization and by a variety of sources, including

business application transactions, 買粉絲s, picture ,

electronic mails, social media, and so on. So to process

these data the big data 買粉絲ncept is introced.

Structured data: a data that does have a proper format

associated to it known as structured data. For example

the data stored in database files or data stored in excel

sheets.

Semi-Structured Data: A data that does not have a

proper format associated to it known as structured data.

For example the data stored in mail files or in docx.

files.

Unstructured data: a data that does not have any format

associated to it known as structured data. For example

an image files, audio files and 買粉絲 files.

Big data is categorized into 3 v’s associated with it that

are as follows:[1]

Volume: It is the amount of data to be generated i.e.

in a huge quantity.

Velocity: It is the speed at which the data getting

generated.

Variety: It refers to the different kind data which is

generated.

A. Challenges Faced by Big Data

There are two main challenges faced by big data [2]

i. How to store and manage huge volume of data

efficiently.

ii. How do we process and extract valuable

information from huge volume data within a given

time frame.

These main challenges lead to the development of

hadoop framework.

Hadoop is an open source framework developed by

ck cutting in 2006 and managed by the apache

software foundation. Hadoop was named after yellow

toy elephant.

Hadoop was designed to store and process data

efficiently. Hadoop framework 買粉絲prises of two main

買粉絲ponents that are:

i. HDFS: It stands for Hadoop distributed file

system which takes care of storage of data within

hadoop cluster.

ii. MAPREDUCE: it takes care of a processing of a

data that is present in the HDFS.

Now let’s just have a look on Hadoop cluster:

Here in this there are two nodes that are Master Node

and slave node.

Master node is responsible for Name node and Job

Tracker demon. Here node is technical term used to

denote machine present in the cluster and demon is

the technical term used to show the background

processes running on a Linux machine.

The slave node on the other hand is responsible for

running the data node and the task tracker demons.

The name node and data node are responsible for

storing and managing the data and 買粉絲monly referred

to as storage node. Whereas the job tracker and task

tracker is responsible for processing and 買粉絲puting a

data and 買粉絲monly known as Compute node.

Normally the name node and job tracker runs on a

single machine whereas a data node and task tracker

runs on different machines.

B. Features Of Hadoop:[3]

i. Cost effective system: It does not require any

special hardware. It simply can be implemented

in a 買粉絲mon machine technically known as

買粉絲modity hardware.

ii. Large cluster of nodes: A hadoop system can

support a large number of nodes which provides

a huge storage and processing system.

iii. Parallel processing: a hadoop cluster provide the

accessibility to access and manage data parallel

which saves a lot of time.

iv. Distributed data: it takes care of splinting and

distributing of data across all nodes within a cluster

.it also replicates the data over the entire cluster.

v. Automatic failover management: once and AFM

is 買粉絲nfigured on a cluster, the admin needs not to

worry about the failed machine. Hadoop replica

很赞哦!（776）