A. Adithyan - Cloud Computing
A. Adithyan - Cloud Computing
Deployment model
Type of service
I.INTRODUCTION
Cloud computing refers to the act of II.DEPLOYMENT MODELS
processing a user given data on a server Cloud computing can be deployed in these
located elsewhere with the use of methods
internet. The idea of cloud computing
started around the 1960s. In this internet Public
connected time period, cloud computing Private
makes a big difference for devices that Community
don’t have heavy processing power built Hybrid
in. Since the processing is offloaded to the Distributed
nearest server farm located somewhere in Multicloud
the world, data can be processed faster Big Data cloud
than ever before. By the use of cloud HPC cloud
computing, heavy processing devices
aren’t necessary to be mass produced.
The operating expenses of a server farm is Public:
high when compared to singular powerful
The resources in this type of deployment
devices. But while serving many users it
is provided for free or at cost to the
becomes relatively cheap
general public for use. This reduces the
Many tech giants make use of cloud wastage of resources as the customer
computing according to their needs. pays for only what they need. It also helps
to serve many common user demands companies keep half their data in the
such as storage in a larger scale. Most cloud and some on premises. It is mostly
organizations try to have their own server also used to overcome the cons of one’s
farm if the data is confidential or if it isn’t deployment with the other. If the
cost effective. company need to comply with certain
security demands
Microsoft’s Azure, Amazon’s AWS are
some dependent companies of the public Some parts of Google, Amazon and
for cloud computing services. Microsoft make use of hybrid type of
deployment.
Private:
Distributed:
This has relatively the same set of benefits
as public, but private hardware is used. The servers are distributed in different
The cost of establishing a private server locations and are connected through a
farm is very expensive compared to using single network. It is of two types Public-
services provided by other companies. resource computing and Volunteer cloud
The data that needs to be processed will computing.
be kept within the companies making the
company more trustworthy.
Multicloud:
Hewlett Packard Enterprise (HPE),
VMware, Dell, Oracle are some of the To reduce the dependency on a single
leading Private cloud providers. cloud computing solution, multicloud is
used. The data that needs to be processed
gets passed through the servers that are
Community: active during the time of request.
Community cloud are designed for
business and organizations that work on a
Big Data cloud:
common project which requires a
centralized storage and computing To analyse and perform operations on
solution this type of deployment is done. large amounts of data, big data cloud is
It does a multi-tenant service. It also had a used. The servers used in this type have a
lot of flexibility as all users of service can large amount of memory to store the
modify it as they wanted it to be. data. They also have more computing
power to serve these needs.
There aren’t many reputable companies
that use this type of deployment as it has IBM, Amazon, Google are some
many security concerns. companies that has this infrastructure.
BaaS:
IaaS:
Most web apps have the basic capability
The companies that render this type of
of presenting the information received
service gives the most crucial things such
from the server. The major of task of
as storage, networking which are
collecting information at one place is done
necessary for a company to be a cloud
by the servers. This helps in the reduction
based one. This has the highest level of
of processing time and will also help
flexibility and is easy to manage. It is most
developers to push updates to their
suitable for temporary workloads.
service for everyone at the same time.
AWS, Digital Ocean, Azure offer IaaS. Bugs and errors are much lower and can
be fixed easily.
Amazon, IBM, HP, SAP and Oracle are
PaaS:
some popular service providers of this
type. Microsoft was one of the first a video to a smartphone, automated
vendors to offer this type of service. vehicle operation to performing surgeries
using computing algorithms in real-time.
There is still room for improvement in
FaaS: terms of power efficiency and flexibility
with cloud computing. There are still
FaaS is similar to PaaS. The major
many countries that don’t have access to
difference is that the developers that use
the internet. The more issues we fix, the
this type of service gets abstracted from
more potential we will get out of cloud
the servers that their project is deployed
computing.
on. Payment are based on consumption
only and not based on server size. This is
mainly useful for running single functions
that are not often used.
Google Cloud Functions, Azure Functions,
AWS, webtask.io are some companies that
offer this type of service.
Serverless computing:
Serverless computing doesn’t exactly
mean that the computing is done without
a server, it is in which the cloud provider
completely manages the virtual machines
required to run the code. It is named
serverless as the process of owning or
renting a server is not done from the
developer’s point of view.
Pivotal software was the first to offer this
type of service. AWS Lambda, Google
Cloud Functions, OpenWhisk are some
companies that make use of serverless
computing.
Conclusion:
Cloud-based technology will continue
expanding even further and may even
eliminate the need of powerful physical
hardware at the hands of each end user.
The potential for cloud computing is
almost limit less. It ranges from streaming