3. Introduction
“Big Data is the next big thing in computing and generates value from very large datasets,
but cannot be analyzed with traditional computing techniques”.
Managing and analyzing data of the customers have been a great benefit for the industries,
but also a great challenge for them. Till the time they were having a handful of customers
things were quite simple, but as soon as the number increased, complications in managing
information raised along with.
Things were not confined to industries only, problems kept arising in medical science,
research and development field, 3D simulations and designing etc. The large data from
medical science gets deleted every week due to insufficient storage management. To face
these challenges the term Big Data came into the scene. ‘Technically speaking, the process of
handling big data encompasses collection, storage, transportation and exploitation.
( who are the generators of Big Data ?)
Some Real Facts:
New York stock exchange generates a 1 TB / day.
Google processes 700PB/month
Facebook hosts 10 billion photos taking 1PB of storage
3
5. 5
Data centers
Such a big data management definitely requires a big big big data center for the
storage purpose. The idea that cloud computing means data isn’t stored on
computer hardware isn’t accurate. Your data may not be on your local machine,
but it has to be housed on physical drives somewhere -- in a data center
Some of the words about Data Centers were given by the experts are given below:-
Data centers are the brains of the Internet.
The engine of the internet, it is a giant building with a lot of power, a lot of
cooling and a lot of computers.
With its row upon row upon row upon row of machines, all working together
to provide the services that make all functions.
6. 6
Management of Data Centers Through
Open source software
Design for large amount of structured, unstructured data
Implemented on the racks of commodity servers as a adopt cluster.
It is designed to parallelize data processing across computing nodes to
speed computation and hide latency.
Hadoop distributed file system
Motivation (Store the data on multiple machines)
Architecture and Concepts (Run on commodity hardware)
Inside (Software is intelligent enough to handle hardware failure)
User Interface (Replicate the data)
Hadoop distributed file system service includes
Namenodes
Datanodes
7. 7
Conclusion
Increase information explosion in every seconds indicates both Big Data
challenge and Big Data opportunity . And the biggest problem rely is the
storage of such data. The storage of your laptop could fit in your hand, but here
they need something bigger. This is where the concept of data centers comes
into picture.
The data centers work and compile all that information and send it back to the
open internet again and all of this happens in milliseconds.
Some people consider the internet a cloud as it is floating around in the sky, but
it’s not, it’s a real physical thing, internet is a real physical building
interconnected with miles and miles of fibers and all of these buildings can talk
to each other and share data back and forth.
These buildings are going to increase day after the day, so we should get ready
to face several new upcoming technologies emerging in the near future.