See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267250165

Cloud Computing: Past, Current and Future

Article · April 2012

CITATIONS READS
0 835

3 authors, including:

Hamid Mcheick Haidar Safa

University of Québec in Chicoutimi American University of Beirut
157 PUBLICATIONS   377 CITATIONS    100 PUBLICATIONS   941 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Undergraduate Project View project

QoS Pattern to Solve SLA Violation During Adaptation Activity of Application Context-aware. View project

All content following this page was uploaded by Hamid Mcheick on 04 November 2014.

The user has requested enhancement of the downloaded file.

 Cloud Computing: Past, Current and Future
H. Mcheick1, F. Obaid2, and H. Safa3
1
Department of Computer Science, University of Quebec at Chicoutimi, Chicoutimi, Quebec, Canada
2
Department of Computer Science, Lebanese University, Beirut, Lebanon
3
Department of Computer Science, American University of Beirut, Beirut, Lebanon

recommend that “developers would be wise to design their

Abstract - With the augmentation of data, the organizations
next generation of systems to be deployed into Cloud
need to provide better and faster secured data access to
Computing”. While many of the predictions may be cloud
customers. Cloud computing came as a godfather of the new
hype, some believe the new IT procurement model offered by
generation of data warehousing and data access through the
cloud computing is here to stay [14]. Whether adoption
network. This new way of deploying and using data solved
becomes as prevalent and deep as some forecast will depend
many of data problems. However, as any new technologies, it
largely on overcoming fears of the cloud.
has its own issues and needs. In this article we are going to
CC represents a new way in how to use and think about
discuss and survey briefly this subject to make an overview of
computing resources and services. Hence, clouds represent a
what might be the past, current and future of cloud computing.
new way in which users store, access, and utilize data and
This might be at the same time the future of information
high-powered resources (IT infrastructure, software and
systems around the world and the future of many
services) remotely via Internet without the need to build large
organizations.
on-site IT systems. The “cloud” is a network of data centers
Keywords: Cloud Computing; Data Access; Data providing commoditized computing power to end users
Deployment; distributed and localized Data. portioned in an as-needed basis to clients in a fashion similar
to electricity or water [2].
This gives individuals and small companies the
1 Introduction opportunity to access vast computing resources with high-
speed, high-quality, and non-discriminatory networks they
1.1 Issues couldn’t use before. Nevertheless, the complexity of this new
technology puts pressure on the existing system: First, the use
When it rains, in few minutes we can see the streets full of distributed systems would change to the use of clouds, so
of water. What gives clouds the ability to do such a seen in all distributed systems need to be changed and companies that
such a time can be easily expressed by the following make these systems would lose customers and money, unless
comparison: The difference between a raining cloud and a these companies begin making cloud systems.
hosepipe is that a cloud drops water in parallel. From this, Second, the computers, networks, and many other components
comes the idea of cloud computing (CC), we don't have to (software and hardware) need to be developed to meet the
have many servers to make parallel transactions [1]. We can needs of cloud systems.
make them from the same site, which will ease the update Thus CC represents a disruptive innovation. It is a new
time and do us big favour in many ways that we'll see in this alternative to acquiring software, platforms, and IT
article. infrastructure as your business grows. Services are now
Cloud computing is a model for enabling convenient, on- accessed over the web on demand from any location via a
demand network access to a shared pool of configurable broadband framework [2]. The CC technology has many
computing resources (e.g., networks, servers, storage, challenges such as failures detection, scalability, security,
applications, and services) that can be rapidly provisioned and specific legislations and licensing models, etc. For example,
released with minimal management effort or service provider the new threats require new constructions to maintain and
interaction [13]. improve security. These challenges will be discussed later in
Cloud computing is the most popular notion in IT today; the article.
even an academic report from UC Berkeley says “Cloud
Computing is likely to have the same impact on software that 1.2 Main Aspects of Cloud Computing
foundries have had on the hardware industry.” They go on to
 Figure-1: Main aspects forming a cloud computing system [3]

o Public cloud: Companies migrate their

infrastructure to an existing public cloud.
The main aspects of CC are:
o Private cloud: Companies build their own
clouds or participate in an interagency
• Types [3]: cloud.
o Community cloud. The cloud infrastructure
o Infrastructure-as-a service (IaaS): it is shared by several organizations and
replaces a company's entire server room supports a specific community that has
and network through virtualization shared concerns (e.g., mission, security
technology in order to cut costs and requirements, policy, and compliance
improve flexibility. Examples: Amazon considerations). It may be managed by the
Web Services, Rackspace, Savvis, HP, organizations or a third party and may exist
IBM, Sun and Google Base. on premise or off premise.
o Hybrid cloud: Companies build their own
o Platform-as-a-Service (PaaS): Delivery of clouds to handle its main applications and
a computing platform and software stack use public clouds to support low-priority
over the Internet to develop and deploy applications.
applications without the necessity to • Benefits: i) Cost reduction: Adapt to changing
manage their own hardware and software consumer behaviour and reduce cost for
layers on-site. For example, Amazon infrastructure maintenance and acquisition, and ii)
Elastic Cloud [15], EMC Atmos [16], Ease of use: Through hiding the complexity of the
Aptana [17] and GoGrid [18] are providing infrastructure.
these services preventing users the
mammoth costs of buying hardware,
• Locality: Local, remote, and distributed.
software and related technology as well as,
to maintain and supporting their IT
infrastructures. • Compared to: SOA, Internet of services, and Grid

o Stakeholders:Users, Adopters, Resellers,

o Software-as-a-Service (SaaS): Provide
and Providers.
online email applications, free services,
limitless storage, and remote access from
any computer or device with an Internet 2 Market Model of Cloud Computing
connection. Example: online email
(Google, Yahoo and Microsoft). 2.1 Competiveness of CC
Since nowadays computing power is cheaper, the
• Features [3]: Elasticity, Reliability, and increase in revenues, customer satisfaction and market-share
Virtualization is becoming obviously clear [3].

• Modes [3]: Its competitiveness is associated to non-functional

capabilities that represent qualities and properties of a system,
 economic considerations, and technological challenges that o Data Management for storage clouds,
arise from realizing non-functional and economic aspects [3]. where data is distributed across multiple
resources.
• Economic aspects [3]: o Metering of any kind of resources /
services is a condition for the cloud
o Reduce the cost of the cloud system that elasticity.
takes into consideration the scalability and o Tools to support development, adaptation,
pay per use (building cost according to and usage of cloud services.
resources consumption) aspects. • Non-Functional Capabilities [3]:
o Reduce the time to market for small and
medium enterprises. o Elasticity: capability to adapt to changing,
o Companies will spend money on dynamic integration.
operational expenditure instead of capital o Reliability:
expenditure that is required to build its own o Quality of Service
infrastructure. o Agility and adaptability: On-time reaction
• Technological aspects [3]: and adaption to changes.
o Availability: The ability to assure
o Virtualization: Virtual infrastructure. redundancy for services and data.
o Multi-tenancy: Implies potential issues. Porter's Five Forces. The figure 2 shows five forces of CC.
o Security, Privacy and Compliance: dealing These forces are described as follows.
with sensitive data and code.

Figure-2: Porter's five forces (Market model) [4]

• New Market Entrants: Many immature entrants that have not Amazon Web Services has since grown into a new
had the time to acquire deep business expertise, is expected business that encompasses compute and storage infrastructure
which will present consumers with a significant challenge in services and new middleware services that Amazon EC2
customers can leverage, such as the Amazon Simple Queue
identifying suitable implementation partners [5].
Service. [6]
• Suppliers: Suppliers are the main player in the CC market, Many followed Amazon to this new IT outsourcing
they own and operate CC systems to deliver services model, The early entrants to CC come from two camps:
(Software, Platform or Infrastructure) to consumers. [5] – Internet services companies. The first camp (led by Amazon
• Buyers (Consumers): The consumers will set the standards in Web Services, Salesforce.com, and Akamai) is turning its
the industry and partly drive the aggregation of players infrastructure management services into profit centers
through more demanding requirements. [5] serving internal and external customers [6].
• Technology Development: Experts agree that the influence of– Hosting providers. The second camp is the forward-thinking
regulators will be a crucial factor in the coming years. hosting providers that see clouds as the next step in the
• Cloud Market: The cloud system itself represented as a market evolution of their business models. This group is comprised
[1]. of large multinationals like Terremark and smaller players,
such as Layered Technologies and XCalibre that see clouds
2.2 Entrants and Networks
as new offerings that can differentiate their businesses.
The first time CC was to be introduced to the world was Startups like Enki, from NetSuite’s data center pioneer Dave
through Amazon Elastic Compute Cloud (Amazon EC2). [6] Durkee, are basing their business on this model [6].
 Many hosting providers (and several enterprises) are
building clouds using 3Tera AppLogic, a grid engine that has Hence, a few large data centers with clouds are likely to
evolved into full-blown CC infrastructure software. This be more ‘green’ than millions of smaller but already large data
technology works across physical and virtualized servers and centers. For these reasons, CC is sometimes referred to as a
has built-in high availability, virtualized storage, and per-use Green information technology as it significantly reduces the
reporting. Unlike most other grid offerings, AppLogic works carbon footprint [10].
with most any type of application without redesign or
programming to a grid API [6].

3 Current and Future of CC

3.1 Current Status of Clouds, the S-Curve
Bandwidth, perception, loss of control, trust and
feasibility were the challenges that confronted the presence of
CC service in the past. Many of these challenges were
overcome by our new technologies and others will be in the
future. Which means that this service moved from virtual to
real and will be advanced as our technologies advance. [7]
The desire to reduce costs and add flexibility to huge
enterprises are the most effective reasons that will make CC
commonly used. [8]
In the coming few years, we would expect more and more
companies to adopt for the cloud solution. Many companies
both big and small, are currently seriously considering
shifting to cloud services because of the many benefits to a
cloud solution. The current status of CC is past the Innovators
stage and gently moving up the lower cusp of the S-curve into
Early Adopter stage (Figure 3) [9].
Figure-3: Current status of cloud computing on the technology
The advantages and benefits of CC are categorized in maturity curve [7]
three categories:
– Centralization: [7] 3.2 Future and R&D
• Competitive advantage in data access. In the future, more cloud adoption is inevitable. However,
• Huge flexibility in data access. for a better assessment of the future of clouds, we will discuss
– Cost: Few huge clouds cost less than thousands some of the current challenges and gaps that are hindering the
rapid evolution of this technology. Such gaps and challenges
of large local servers. Less materials, less areas, can be divided into two categories: technical and non-
fewer employers. technical. After discussing the gaps, we will suggest
– Environmental effects: [7] i) Less need for corresponding set of suitable R&D steps that needs to be
infrastructure, ii) Less need for hardware, iii) taken to overcome these challenge and make a better use of
the CC system (Attention: some gaps might happen in the
Huge reduction in energy consumption. future and will need to be taken care of). [3]
The advantages and benefits of CC are categorized in– Technical: [3]
three categories: 1. The ability to detect failures, adapt to the required scale
– Centralization: [7] of resources., ensuring continuous availability of such
• Competitive advantage in data access. resources, and meeting clients expectations in terms of
• Huge flexibility in data access. quality.
– Cost: 2. Privacy and Security.
Few huge clouds cost less than thousands of 3. New security holes will appear with hackers advancing in
large local servers. Less materials, less areas, fewer their efforts.
employers. 4. Adaptability. Example: If a CPU is added to a virtual
– Environmental effects: [7] machine that is already in use, the running code should be
• Less need for infrastructure. able to adapt and make use of the additional resource
• Less need for hardware. without having to be restarted or even adapted.
• Huge reduction in energy consumption. – Non-Technical: [3]
 5. All data, resources, applications, and services need
specific legislations and licensing models depending on
the location they are hosted in. The challenge with such a
step is the different laws put forward by different
countries which might make it hard for hosting such a
service in some locations.
6. Huge investment and Intensive research.
– Solutions: [3]
1. New segmentation concepts and distributed programming
models.
2. Solution: Data should be simultaneously protected in a
form that addresses legislative issues with respect to data
location and be manageable by the system.
3. Solution: Imposing clear legislation models regarding
jurisdiction over the hosted data and its distribution in
other countries.
4. Solution: Developing programming models to provide
adaptability.
5. Solution: All legislation issues should be properly
addressed by respecting the uniqueness and distinctive Figure-4: Hybrid clouds with private and public cloud models
nature of some countries and playing by their rules. [11]
6. Solution: New expert systems and best use Recommendations. We now see CC as a shift from using
recommendations. many big servers to using one giant server, from distributed
4 Discussion and Recommendations data (of the same company) to centralized, though, it's still to
be distributed when 2 companies agree to share data. For the
Discussion. No matter what your applications and first company, its clouds are private and local, and the second
requirements are, cloud services could present the solution for company's data are public and distributed. For the second
such necessities. Companies might not fully switch to a cloud company, it's vice versa. For example, if Facebook and
service but rather go for partial hybrid switch as it serves their Twitter agree to share information about their users, Twitter
specific needs. Microsoft for example, adopts such a strategy, data are public and distributed for the Facebook company, but
and thus, they have built their cloud solutions based on hybrid these same data are private and localized in Twitter eyes.
(Figure 4) type offerings [11]. Using clouds is to simplify what was complex and ease the
access of what was difficult to access, many other benefits of
As an easy and simplified guideline for companies using clouds, some were mentioned in this article, some
thinking of cloud as a potential solution, we recommend the weren't, and some are to be discovered. [1]
following analysis as a potential startup [11]. We recommend the following:
When considering cloud, one should go into data physics1. Use CC systems whenever possible.
to study the time and cost consumption of such technology.2. For the best use of these systems we recommend spending
Data physics correlates between processing elements and the some time and money on doing researches because it's always
data on which they operate. The fact that CC stores all data in
better to lose a few thousands of dollars than to risk a few
a cloud rather than physical hardware (hard disks) imposes the
issue of time consumption taken by data to move from the millions.
cloud to the relevant server for processing. [11] 3. CC might be a benefit for some companies and a disaster for
The governing equation for time consumed by data to other companies, so when considering cloud computing we
move from its storage location to the processing server is: must do our own researches and not to totally depend on
time = bytes * 8/bandwidth other's researches.
where the bandwidth to and from the cloud provider is the4. Companies should contribute together in the use of clouds by
bottleneck. Therefore, the memory capacity serves as the key
bottleneck for virtual machine density and improving such a sharing "none personal" data, it has almost no disadvantages
key factor plays a major role in the decision of “cloud and could have many benefits.
computing” deployment. [11] 5. CC systems could be more useful, all we need is to upgrade
So as a simple rule of thumb, if the time calculated to our technologies to meet the requirements for such a useful
move data from cloud to server or cloud to cloud is more than system.
the relative data processing time, then implementing such a
strategy becomes worthless [11].
6. Cloud fears largely stem from the perceived loss of control of [5] "Is Cloud Computing Ready For The Enterprise?".
sensitive data. Current control measures do not adequately Available at: http://www.3tera.com/download/Forrester-
address cloud computing’s third-party data storage and Cloud-computing-report080307.pdf
processing needs. New visions propose to extend control
[6] " Cloud Computing; The Past, The Present, The Future
measures from the enterprise into the cloud through the use of
(Part 1)". Available at:
Trusted Computing and applied cryptographic techniques. http://www.windowsecurity.com/articles/Cloud-Computing-
These measures should alleviate much of today’s fear of cloud Past-Present-Future-Part1.html
computing, and, is believed, have the potential to provide
demonstrable business intelligence advantages to cloud [7] "Hardware vs. Software: The Defining Technology
participation. Battle of This Decade". Available at:
http://www.xconomy.com/boston/2010/01/20/hardware-vs-
5 Conclusion software/
Cloud computing (CC) offers an exciting opportunity to
build data structures that promise to solve problems associated [8] " Cloud Computing Climbs Adoption S-Curve".
with economic modeling, terrorism, healthcare and epidemics, Available at:
etc… and to bring on-demand applications to customers in an http://www.informationweek.com/news/software/info_manag
environment of reduced risk and enhanced reliability [1]. ement/228900680?queryText=
Moreover, clouds could play a major role in climate
change as they have proved to be a reliable green option that [9] "Cloud computing's green paradox". Available at:
will contribute to reduction of carbon footprints. CC promises http://news.cnet.com/8301-19413_3-10428065-240.html
to reduce run time and response time of deploying
applications, increase the pace of innovation, and lower entry [10] "INTRODUCTION TO CLOUD COMPUTING
costs, all while increasing business agility. As much as it ARCHITECTURE - White Paper - 1st Edition, June 2009".
seems promising, successful deployment of cloud technology Available at:
requires change of design for current existing applications and www.pcworld.com.vn/Handler/FileDowload.ashx?fileId=515
cannot just be unleashed on the cloud as it is [1]. 1
Design enhancements should focus on improving
scalability factors to ensure service could keep up with [11] " Cloud computing leaders". Available at:
demand requirements. If such relevant design upgrading is http://www.cloudcomputingleaders.net/
applied, deploying cloud technology will provide customers
with a unique and genuine opportunity to reap the benefits of [12] Peter Mell and Tim Grance. “Effectively and Securely
achieving the competitive advantage of a flexible, scalable, Using the Cloud Computing Paradigm”. NIST, Information
high speed, error reduced, and environmental friendly Technology Laboratory, 2009.
application solution.
[13] Armbrust,M., Fox,A.,Griffith,R. et al. Above the Cloud:
A Brekely View of Clod Computing. UCB/EECS-2009-28,
6 References EECS Department, University of Califonia, Berkely,2009.
[1] Writers own information, ideas, understanding to the
subject and point of view."Cloud Computing". Available at: [14] http://aws.amazon.com; Accessed on February 21st,
http://www.ccianet.org/CCIA/files/ccLibraryFiles/Filename/0 2011
00000000151/Cloud_Computing.pdf
[15] http://www.emc.com/products/family/atmos.htm;
[2] "THE FUTURE OF CLOUD COMPUTING - Accessed on February 21st, 2011
OPPORTUNITIES FOR EUROPEAN CLOUD
COMPUTING BEYOND 2010". Available at: [16] http://docs.aptana.com/docs/index.php/Aptana_Cloud
http://cordis.europa.eu/fp7/ict/ssai/docs/cloud-report-final.pdf Docs_Wiki; Accessed on February 21st, 2011

[3] "Cloud Computing in Research - SYSTEMATIC [17] http://www.gogrid.com; Accessed on February 21st,
THOUGHT LEADERSHIP FOR INNOVATIVE 2011.
BUSINESS". Available at:
http://www.ccice.org/files/Maik_Lindner.pdf

[4] "Cloud computing Forecasting change". Available at:

https://www.deloitte.com/assets/Dcom-
Global/Local%20Assets/Documents/TMT/cloud_-
_market_overview_and_perspective.pdf

View publication stats