Cloud computing
Introduction
Cloud
computing was and is still a much discussed subject in the field of computing
and the internet. This happens because this technology has a lot of strong
advantages but disadvantages as well. This technology enables centralized
communication with computer resources and data over the network. These
resources including computers, servers, printers, video game consoles, etc are
connected to the wider network infrastructure. Data is usually stored in
servers located in massive data centres. It is not a new concept, as its
general idea is based on an older computing model known as “Timesharing”. It is the result of more than forty years of evolution. Cloud computing nowadays, is
evolving faster than ever before, and day by day, the number of companies using
is increasing. In this article we are going to present both some of its
strengths and weaknesses and how it works.
Cloud structure
The
principal structure of Cloud computing involves two parts which are connected
to each other via a network:
The Front end
In a “Front end” example, a computer acts as a client and
includes the application needed to gain access to the back end (Cloud service).
Such applications are known as APIs (Application Programming Interface). APIs
are protocols used as user interfaces for providing various functions of
software communication.
The
best way to think about cloud computing is “Data Centre with API”. [3]
Different applications use different interfaces. Some
applications enable network connection directly through the web (e.g. e-mail)
and some others through specifically designed network interfaces.
The Back end
The Back end
In a “Back end”
example, the cloud service is communicating with a client’s computer. A cloud contains
servers, computers, network routers, database storage machines and more. Also
there are applications stored there, such as video editing, video gaming, word
processing and anything that can fit in software. These applications can be
accessed by a client computer, via communicating with the corresponding
application’s servers as shown in Figure1.
A control node,
typically a central server is used for administrating and monitoring the whole
network’s operation.
Most of services
provided, are transparently delivered, without the physical server knowing much
about where the data is actually stored. This concept is similar to
virtualization, where multiple resources seem like a single one.
In server environments,
virtualization has become the norm with hypervisors providing for increased
flexibility and improved utilization of physical assets [4].
Storage
Storing
of the data takes place in massive data centers (Figure 2). Data centers house
server machines, which act as database or application storage mediums.
As we know, servers often require maintenance or repair, and
thus it is important to store copies of the data in multiple places. This is
done or should be done to achieve redundancy
of data. Without it, a cloud service provider cannot ensure its
clients anywhere and anytime access of data. Also data is stored in servers
using different power supplies or located in different places, to overcome the sub-sequence of a power failure event.
As
we noted earlier, servers can store both data and applications. Some
applications are web-based and can access the cloud through an internet
browser. Such applications include:
- Web mail
· - Social
networking
· - Multimedia
albums
· - Video
hosting
· - Online
gaming
· - Online
documents
In
the other hand, some applications require the use of specific network enabled
interfaces. Such applications include:
· - Antivirus
software
· - Network
monitoring tools
· - Operating
system/software updating
· - Data
sharing (Drop box)
Advantages of the
Cloud
Cost
The
most important advantage is that it can reduce the cost of the infrastructure,
of a company using it. It is easier to maintain and upgrade, because network infrastructure
is centrally located and can be accessed both remotely and locally.
Storage
Storage
space in a cloud can be continuously expanded, as there is no limit in disk
sizes. Data is hosted in third party servers and it is the host’s
responsibility to take care of such issues. Such data can be accessed via a
web-based interface, e.g. Google’s “drive” feature, which enables saving files
on the cloud.
Backing
up and recovery of data is also easier to be implemented. Most third parties
will keep copies of the data, and thus it will be easier for it, to be
recovered. Furthermore, data is stored in a safer place than a local physical device,
in terms of objective loss or destruction of data. This reliable and redundant
storage of data can provide strong “Business continuity” and “Disaster recovery”.
A minor concern about storage, is that despite that most cloud services try to issue storage concerns via redundancy,
there's still the possibility that an entire system could be destroyed (e.g.
fire, flooding), and data will be lost completely. So it is important not only
to store data in different logical places but in physical too.
Subjective
loss, theft and privacy of this data issues will be discussed later.
Remote access
Data
stored in the cloud, can be accessed from anywhere at any given time, given
that a network connection is provided. This enables geographical and time
independence. An employer of a company can retrieve his office work from
whenever he wants and thus improve his productivity. Carriage of physical
storage devices is not needed, and thus possibility of physically losing it is
not present. Other people can also be enabled to access your data remotely.
Disadvantages of the
Cloud
Security
Cloud
computing means Internet computing [5].
The
biggest disadvantage of the internet today, is security. And the same happens
for the cloud services, as those reside on the internet. Cloud data, is stored
in third party servers and thus is vulnerable to loss or theft. Some hosting
providers can be more reliable than others and hence it is important to choose
the best one. As cloud service’s use
increase, more and more hackers will try to break into the system. This reveals
that strong encryption, decryption and authentication schemes need to be used. RADIUS protocol is an example of this discipline
applied to a server infrastructure, and which provides a decent level of data
protection.
Technical issues
Cloud computing need more
resources, smarter computers and thus more capable processor units. There is
also the need of excellent, 24/7 internet connection. If for any reason
connections fail, the ability to access and upload data will be hugely affected.
Also there will be times where systems will face hardware or software malfunctions,
and if not back-up plans are deployed a serious outage will be caused. A company
served by a cloud service provider, should also keep back-ups of its data
locally too. This again reveals the lack of reliability of the cloud. If so
many measures need to be taken, then what is for?
A serious service outage who I experienced myself too, was
the Playstation network’s one, which lasted for 25 days and affected millions
of users. In this case hackers attacked user accounts and the service had to
face major changes in security [6]. Users pay a lot of money to such services,
and are expecting their information to be kept securely and have 24/7 service
available.
Privacy
Nowadays
tons of user data resides on the internet. If somebody thinks that is not a
cloud service user, then he should think twice. Facebook, twitter, gmail are
only some of the web-applications that are keeping our personal information of
all sorts of things. This data is circulated and stored in unknown third party
storage places. Of course most of these provide service level agreement
contracts that will cover various privacy concerns. Personal data should be
processed in accordance to these contracts. Legal measures will be able to
cover most of these concerns, but in some cases the laws will be applied only
within the country that the company or the service resides. Legalisation of
different areas of IT is very challenging and complicated, and will always
struggle people and companies
Conclusion
Like most IT technologies, we saw that cloud computing too
carries a lot of advantages and disadvantages. While the technology is totally
evolutionary, it can also cause great harm to an amateur user, who do not
understand and use it properly.
Most of people nowadays are unaware that the whole internet
is dominated by cloud services. Therefore I believe that some computing privacy
and security issues should be taught to younger people in schools, because by
that way, they will possess a greater awareness of the dangers that surround
the current and upcoming use of the internet.
From a business perspective, we saw that a very big
opportunity is opened for smaller and bigger IT companies. The evolution of
internet and computing speeds, gives the ability for efficient virtualisation
and remote storage of data, with lots of benefits. It is up to the companies to
fulfil user needs and eliminate the presence of some disadvantages. As internet
security evolves and improves, so cloud computing will become better and more
trusted.
Nikolas Georgiou
Nikolas Georgiou
