ISSN 2229-5984 (P)

2249-5576 (e)

Cloud Computing Research Around the Globe in Different

Universities: An Overview
Mayank Yuvaraj and Ajay Pratap Singh

To Cite: Yuvaraj, M. & Singh, A.P. (2013). Cloud computing research around the globe in different universities:
An overview. International Journal of Information Dissemination and Technology, 3(4), 230-235.

DLIS
ABSTRACT
Banaras Hindu University
The first decade of the twenty-first century has not witnessed a world without
Varanasi
ubiquitous and virtualized cloud computing power. The study analyses the pattern of
research initiatives carried around the globe in different universities as well as briefly
elucidates the notion of Cloud Computing(CC). The study will help the researchers to
define their research problems and understand the concept at its optimum.
Corresponding Author KeyTerms: Cloud computing, Cloud Computing Research, SaaS, PaaS, IaaS, Research
Mayank Yuvaraj Initiatives, Cloud Computing in Universities.
[email protected]

Received On: 08.03.2013; Revised On: 30.11.13; Accepted On: 04.12.13

INTRODUCTION CC has been defined as an integrated package of computing

services and applications on web, offered as a utility ensuring
Tremendous amounts of research effort have been devoted to throughout abstraction shown in figure 1. It provides an
studying different issues and problems of CC. Although, alternative to traditional computing where the operating systems,
researchers have recognized CC as a legitimate area of research hardware and software are rented over the Internet. Unlimited
yet CC research is scattered across a wide range of disciplines and ICT capacity and applications are offered as an online service
published sources. The present paper aims to provide an through CC. Unlike traditional computing environment in a CC
overview of CC research around the globe in different niche the cloud service provider looks after the resources such as:
universities. Overall, researchers have addressed different issues hardware, maintenance, updating, security, and other essential
and problems of CC as well as proposed many solutions, while resources. The customer is independent of IT bothering and he
some are still being investigated. Moreover, it is impossible to has to only find the right services which can enable the customer
provide information about all the research activities that have to improve the computing performance. Although the concept
taken place so far. So, avoiding any disingenuous labor we have was in use long time back but the term was formally announced
put forth the concept as well as the relevant CC research by Google in 2010 which most of the scholars feel as a marketing
initiatives carried out in universities. strategy.
CLOUD COMPUTING: DEATH OF HARDWARE The environment of academic libraries has undergone a drastic
change in the last few years. With the emergence of Internet era
In the recent years Information Technology sector has witnessed
libraries are facing a big question to justify their services.
technological turmoil in the form of thunders of CC. The concept

Fig. 1: Cloud Computing Concept

International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4 230
Moreover, academic libraries have discovered a sharp fall in the  Software as a Service (SaaS)
number of gate counts as well as reduction in the circulation of
In SaaS model, the applications or software is delivered as a
traditional library materials. Some researches indicate that a
service to the end user, who can access the program online using
rapidly growing percentage of the use of electronic resources
a web browser or any other suitable users. SaaS services may be
occurs outside of the library. Where once students and faculty
available on rental basis or on peruse basis. For example email
turned to libraries, they now turn to their personal computers,
and few other services offered by google, hotmail, yahoo, sify,
tablets or smart phones when they need to find information.
skype etc. CC offers the ability of libraries to use online software to
Under such circumstances, CC is an approach to computing that
handle a task like video chat through gmail video chat or through
could be worthwhile to academic libraries. Academic libraries
skype8. Both of these are free services though there is little
can choose to how to get them connected to become a part of the
customization or control available with these applications11.
cloud by opting from variations of software, systems, and
Since the services and application interfaces are often familiar
hardware services. Whatever they choose CC could help the
with users, it will decrease the learning curve for library staff and
academic libraries save time, money and resources if servers and
users8.
software were not needed on their premises as they are today.
Hopefully libraries could focus more directly on services and  Platform as a Service (PaaS)
materials for patrons if their computer hardware and software In PaaS model, a computing platform supplies tools and a
were handled by IT companies of the cloud. There are many development environment to help organizations to build, test,
reasons why CC is being common16. Technologically, we use CC and deploy Web-based applications. It helps organizations not to
because we can and it is convenient to use. Economically, it is make investment in the infrastructure required for building Web
cost effective and pocket friendly, and finally it makes and mobile applications, but can rent the use of platforms. PaaS is
interactivity easier to achieve with targeted audiences. based subscription model so users only pay for what they use.
CHARACTERISTICS OF CLOUD COMPUTING Special platform or application infrastructure is also being
provided to the clients. Moreover, client does not require
 Self-seeking on-demand service in cloud knowing programming language, database management
 Wider accessibility to networks systems, etc. to run applications. Windows azure, google app
engine and force.com are the finest examples of PaaS. Libraries
 Metered service depending on the usage of cloud based can create applications in online environment where they can
services build, test, and deploy web based applications11. PaaS gives the
 Flexibility or resiliency in cloud library the freedom to explore development options without
having to purchase and maintain the required infrastructure8.
 Pooling of resources available in cloud
 Infrastructure as a Service (IaaS)
 Virtualization of computers
In IaaS model, computer infrastructure is offered as a platform in a
 Green Computing
virtualization environment which can be used as a service along
 Means to deliver IT to the end-users. with raw storage and networking. IaaS is priced on a pay-as-you-
go model enabling clients to scale up or down the operations
 Flairs of Cloud Computing
depending on their needs at any given time and pay only for what
Cloud Computing offers various services as shown in figure 2 they use. In the simple way, hardware services such as processors,
which has been illustrated below: memory, networks etc on agreed basis for specific duration and
price. In IaaS, clients are being offered with storage, networking
and processing of data. Amazon's Elastic Compute Cloud (EC2) or
Simple Storage Service (S3), VMWare vCloud are some of its
examples. By using IaaS, a library can purchase server space and
computing power8. A library does not need to purchase a server
which is underutilized but costs the same to purchase and
maintain as if it were using all of its resources at all times. By using
an IaaS, a library gains the benefit of only paying for the resources
you actually use14.
Any organizations can deploy CC applications on public, private,
hybrid and community clouds model.
Craggs6 has put forth the main categories of CC deployment:
 The Public Cloud: IT resources and services are owned by a
third party, located off-premise and made available to
anyone on a commercial basis as metered services.
 The Community Cloud: IT resources and services are owned
and operated on behalf of a community of organizations.
 The Private Cloud: IT resources and services are
owned/leased by a single company for its own use.
Fig. 2: Flairs of CC

231 International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4
 The Internal Cloud: A private cloud where all resources Researchers at CMU are using CC for multi-tier indexing of Web
remain on-premise. search engines. They are using CC to characterize the topicality of
web content to more effectively process web searches. Routing
 The Hybrid Cloud: A combination of two or more cloud
searches topically requires less effort than traditional searches,
models.
enabling significant computational and financial savings. The
 Research Initiatives in Cloud Computing project is using the Google/IBM cluster to 'crawl' the web and
Research initiatives are an inventory of research processes perform the data cleansing and pre-processing necessary to
undertaken by various institutions to solve a research problem. develop a web dataset of 1 billion documents to support the
Some of the notable research initiatives to the cause of CC is research. The web dataset is also being made available to the
enlisted below: larger information retrieval community to multiply the impact of
the project on that discipline.
 Boston University (BU) Cloud Initiative, U.S.A.
 Duke University (DU) Cloud Initiative
Funded by National Science Foundation grants BU has taken
initiatives in distributing/CC that aims to offer an open Funded by National Science Foundation in collaboration with
marketplace to independent, rational parties interested in setting NCSU, UNC Chapel Hill, and NCAT State University to explore
up their own applications. They are exploring the competencies and test Trustworthy Virtual CC Duke University has created a
of Colocation Games5 to develop an economically-sound trustworthy virtual cloud super-computing environment for
framework upon which emerging cloud architecture could be researchers. DUs Scalable Computing Support Center2 has
implemented. developed two options for CC:

 Carnegie Mellon University (CMU) Cloud Initiative, U.S.A.  Private Duke Cloud, where researchers can purchase
high-priority cycles as needed from a pool of machines
CMU5,1 is host to several CC research programs. CMU has joined and pay per CPU hour.
hands with silicon valley based researchers to address the need
for industry-wide, globally accepted measures for calculating the  Economy Cloud, where researchers can purchase low-
benefits and risks of CC services. Development of standards and priority cycles only, for a lower per CPU rate.
measures to determine the cost, risks, quality and performance for Researchers pool existing resources and share access to greater
CC services is also on top priority. Further, CMU in Qatar along computing power than they could afford on their own, without
with IBM (International Business Machines Corporation) is worrying about the 'care and feeding' of the equipment, since DU
working on Cloud Computing Lab4 (CCL) research initiative from provides the physical space, systems administration,
2009 to bring CC to the middle east. It focuses on performance programming assistance, power and cooling for the machines.
analysis of scientific workloads, with regional relevance, on the
 Florida International University (FIU) Cloud Initiative
cloud as well as solutions to overcome current limitations of the
existing cloud paradigm. FIU7 researchers are leveraging CC to analyze aerial images and
objects to help support disaster mitigation and environmental
Also, Intel Science and Technology Center for Cloud
protection. Prominent researches carried in the CC domain
Computing10 (ISTC-CC) has been formed as an open community
include creation of a virtualized infrastructure systems and
to devise critical new underlying technologies for the cloud
applications laboratory and Verification based Integrity
computing of the future. It is headquartered at CMU that includes
Assurance Framework (VIAF).
researchers from GeorgiaTech, Intel, Princeton and UC-Berkeley.
They focus on future clouds and its applications that is carved on  Indian Institute of Technology (IIT) Cloud Initiative, Delhi
four inter related pillars shown in figure 3.
In India, IIT has played a pioneering role and is the earliest
adopter of CC solutions. IIT Delhi has developed Baadal in 2011,
which is an indigenously developed cloud orchestration and
virtualization management software that can work with multiple
virtualization technologies like KVM, Xen and VmWare. Further,
IIT Delhi has commissioned ownCloud, a file and document
sharing utility similar to the popular dropbox, for use by the IIT
Delhi community. The utility supports storing and sharing of files,
images, music and documents, contacts, calendar, tasks etc. It
also supports version control and syncing with
Windows/Linux/Mac desktops and Android and iOS based
devices.
Baadal is still in its beta phase which is based on:
 32 blade servers each with 2x6 core Intel(R) Xeon(R)
CPU X5670 @ 2.93GHz and 16 GB RAM.
 16 blade servers each with 2x4 core Intel(R) Xeon(R)
CPU E5540 @ 2.53GHz and 12 GB RAM
Fig. 3: Focus Areas of Intel Science and
Technology Center for CC  A 10Gbps ethernet backbone

International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4 232
  50 TB of virtualized storage based on a NetApp 3210V The VCL solution allows users to remotely access a desired set of
applications and environments over the Internet using a personal
NAS and HP EVA6400 SAN with FC disks.
computer, laptop or mobile device from anywhere, at any time.
 Open source virtualization technology based on KVM. To VCL users, even the most demanding software applications,
 Indiana University (IU) Cloud Initiative operating systems and environments are easily accessible
through license-honoring technology in a click of the mouse.
The researchers at IU are working on several CC projects with Access is instant, and offers a range of options from single
grants from the National Science Foundation (NSF) and the desktops to classroom-sized labs, to collections of servers and
National Institute of Health (NIH). Their research includes Large- storage, to high-performance computing clusters.
Scale Distributed Scientific Experiments on Shared Substrate15,
Exploring the use of cloud techniques to overcome current Further, NCSU13 is engaged in several CC projects funded by the
medical computing obstacles such as long computation time and NSF which includes two collaborative studies on Trustworthy
large memory requirements; and The FutureGrid project that will Virtual CC and Hybrid Opportunistic Computing for Green
provide an experimental platform that accommodates batch, grid Clouds.
and CC.  Purdue University (PU) Cloud Initiative
 Massachusetts Institute of Technology (MIT) Cloud Initiative The project at Purdue University is investigating linguistic
MIT 5 in collaboration with Yale University (YU)and the extensions to MapReduce abstractions for programming modern,
University of Wisconsin at Madison are working on comparative large-scale systems, with special focus on applications that
study of approaches to cluster-based, large-scale data analysis manipulate large, unstructured graphs. This will impact a broad
funded by NSF. In addition they are also independently studying class of scientific applications. Graphs have important utility in
CC Infrastructure and Technology for Education. the social sciences (social networks), recommender systems, and
business and finance (networks of transactions), among others.
 North Carolina Agricultural & Technical State University The specific case study targeted by the research is a comparative
(NCATSU) Cloud Initiative analysis of graph-structured biochemical networks and pathways
The team at the NCATSU is conducting research, funded by a which underlie many important problems in biology. They are
NSF, in collaboration with NCSU, DU, and the University of NC also providing a CC testbed called Wispy to TeraGrid users.
at Chapel Hill to explore and test Trustworthy virtual CC.  Tsinghua University (TU) Cloud Initiative, China
 North Carolina State University (NCSU) Cloud Initiative TU3 is the first prestigious university in mainland China to join
NCSU and IBM have developed VCL (Virtual Computing Lab) Google's CC related programs and to work with Google on the
shown in figure 4 to provide every student with most advanced offering of data processing courses and the research on CC. They
educational resources. Through cloud infrastructure students will have developed TU Cloud which is an all in one CC solution
be provided accessibility to most advanced educational developed at the Grid Computing Division of TU in China that
materials, select software applications and computing and comprises of three components: Nova (virtual computation
storage resources. system: computing cloud), Carrier (distributed file system) and
Corsair (distributed file manager based on carrier: storage cloud),
which can be utilized independently or in combination.

Fig. 4: Virtual CC lab of NCSU

233 International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4
 University of California (UC) Cloud Initiative, Irvine  University of Virginia (UV) Cloud Initiative
Funded by NSF, UC Irvine12 is working on a project to provide The team at the UV is working on several CC projects funded by
support for efficient fuzzy queries on large text repositories in the NSF. They include, feedback controlled management of
cloud. Supporting fuzzy queries can ultimately help applications virtualized resources for predictable escience and image super-
mitigate their data quality issues because entities with different resolution using trillions of examples.
representations can be matched for example 'PO Box' vs 'P.O.
 University of Washington (UW) Cloud Initiative
Box'.
The researchers5 at the UW are working jointly with University of
 University of California (UC) Cloud Initiative, San Diego
Utah on building a new infrastructure for computational
Researchers at the UC, San Diego are studying how to manage oceanography that uses the Google/IBM cloud to allow ad hoc,
and process massive spatial data sets on large-scale compute longitudinal query and visualization of massive ocean simulation
clusters. This research will use the LiDAR topography data hosted results at interactive speeds. Further, their Astronomy Survey
by OpenTopography as a test case and will focus on how CC can Group is conducting research to scale the sky with
aid the management and processing of massive spatial data sets. MapReduce/Hadoop.
 University of California (UC) Cloud Initiative, Sant Barbara  University of Wisconsin (UW) Cloud Initiative, Madison
The UC, Sant Barbara is actively pursuing several advancements The UW, Madison have designed the Hierarchically Redundant,
in CC. Their Massive Graphs in Clusters (MAGIC) project is Decoupled storage project (HaRD) to investigate the next
focused on developing software infrastructure that can efficiently generation of storage software for hybrid Flash/disk storage
answer queries on extremely large graph datasets. They have also clusters. They are also working with MIT and Yale University (YU)
designed an open-source implementation of the Google on a comparative study, funded by a NSF CLuE grant (Cluster
AppEngine interface. Exploratory Grant), of approaches to cluster-based, large-scale
data analysis.
 University of Maryland (UM) Cloud Initiative, College Park
 Wayne State University (WSU) Cloud Initiative
The research team at the UM, Cloud Computing Center at College
Park is working on a range of projects funded by the NSF. They Wayne State University is working on developing a unified
include a Hadoop Toolkit for Distributed Text Retrieval, Data learning approach, namely URL, to automate the configuration
Intensive Text Processing, Commodity Computing in Genomic processes of virtualized machines and applications running on
Research, and a series of other independent studies. the virtual machines and adapt the systems configuration to the
dynamics of cloud. Further, they have started Wireless health
 University of Massachusetts (UM) Cloud Initiative
initiatives 17 that attempts at applying sensors, wireless
CC research in University of Massachusetts involve the communications and CC to healthcare applications.
Google/IBM cloud to learn more about world relationships which
 Yale University (YU) Cloud Initiative
is funded by NSF.
The team at YU is working in collaboration with MIT and the UW
 University of Melbourne (UM) Cloud Initiative
at Madison on a comparative study, funded by a NSF, CLuE grant,
The Cloud Computing and distributed systems laboratory at UM of approaches to cluster based, large-scale data analysis.
is actively engaged in the design and development of next-
generation computing systems and applications that aggregate or CONCLUSION
lease services of distributed resources depending on their Cloud Computing is being considered as the future of computing
availability, capability, performance, cost, and users' quality-of- technology. From the study it is clear that there is an acute dearth
science requirements. The lab is working towards realizing this of research initiatives in developing countries like India.
vision through its two flagship projects: Gridbus and Cloudbus. Although it has many merits and is an end to painless works of
 University of Minnesota (UM) Cloud Initiative updating, configuring and installing software educational
organizations should look forward for CC solutions. The
Researchers at UM are developing a cloud proxy network that
implementation of CC will not be succeeded if universities
allows optimized and reliable data-centric operations to be
merely view as an online applications. Although, some of the
performed at strategic network locations.
universities have got discovery services, e-mail service on cloud
 University of North Carolina (UNC) Cloud Initiative, Chapel and are further moving ahead step-by-step sharing resources
Hill among libraries however moving in patches to cloud is risky and
not going to be fruitful in the long run. There is a need of proper
In collaboration with NCSU, DU, and NCAT State University,
training and plan well before introducing CC in the educational
researchers at the UNC are trying to explore and test trustworthy
setup supplemented by well planned research initiatives.
virtual CC environment.
 University of Utah (UU) Cloud Initiative REFERENCES
The researchers5 at the UU are working jointly with University of 1. A break in the cloud. Qatar initiatives open doors. Carnegie
Washington(UW) on building a new infrastructure for M e l l o n U n i v e r s i t y. ( A c c e s s e d o n 0 2 . 0 8 . 2 0 1 2 ) ,
computational oceanography that uses the Google/IBM cloud to http://www.cmu.edu/homepage/society/2009/winter/a-
allow ad hoc, longitudinal query and visualization of massive break-in-the-clouds.shtml
ocean simulation results at interactive speeds.

International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4 234
2. Bonnett, C. (2011). Venturing into the computing grid. 11. Kroski, E. (2009). Library cloud atlas: A guide to cloud
Duke Today. (Accessed on 13.08.2012), http://today. computing and storage stacking the tech . Library
duke.edu/2011/05/dukecloud Journal.Com. (Accessed on 14.08.2012), http://www.
libraryjournal.com/article/CA66957 7 2.html
3. Chinatechnews.com Google to work with Tsinghua
University on Cloud Computing (2008). (Accessed on 12. National Science Foundation. Press Release 09-082
09.08.2012), http://www.chinatechnews.com/2008/03/18/ National Science Foundation Awards Millions to Fourteen
6503-google-to-work-with-tsinghua-university-on-cloud- Universities for Cloud Computing Research. (Accessed on
computing 02.08.2012), http://www.nsf.gov/news/news_summ.jsp?
cntn_id=114686
4. Cloud Computing Lab. (Accessed on 02.08.2012),
http://ccl.qatar.cmu.edu/ 13. NC State University Computer Science. CSC News (2008).
North Carolina State University and IBM extend access to
5. Cloudbook: Telling the cloud story. Research Clouds.
educational resources to the world through Cloud
(Accessed on 06.08.2012), http://www.cloudbook.net/
Computing. (Accessed on 16.08.2012), http://
directories/research-clouds/cloud-computing-research.php
www.csc.ncsu.edu/news/764
6. Craggs, S. (2009). Cloud computing without the hype: An
14. Peters, C. (2010). What is cloud computing and how will it
executive guide. (Accessed on 12.08.2012), http://
affect libraries? Techsoup for Libraries. (Accessed on
www.slideshare.net/LustratusREPAMA/cloud-computing-
12.08.2012), http://www.techsoupforlibraries.org/
without-the-hype-an-executive-guide-100-slideshare
blog/what-is-cloud-computing-and-how-will-it-affect-
7. FIU School of Computing and Information Sciences libraries.
brochures. (Accessed on 10.08.2012), www.cis.fiu.edu/
15. Researches in Cloud Computing. (Accessed on 07.08.2012),
brochure/SCISBrochure2012FINAL.pdf
http://www.the-cloudcomputing.com/researches-in-cloud-
8. Goodman, A. (2011). Cloud Computing in Libraries and computing.aspx#
Library School . (Accessed on 04.08.2012), http://www.
16. Sachdeva, M., Rana, P., Kapoor, R. & Shahid, M. (2011).
amandagoodman.com/cloud-computing-in-libraries-and-
Cloud computing pay as you go technology. Proceedings of
library-school/
the 5th National Conference on Computing for National
9. IBM Corporation (2010). (Accessed on 05.08.2012), Development, 161. New Delhi: BVICAM.
http://www.ibm.com/ibm/files/K6403W72867H78/12
17. WHI: Wireless Health Initiative at Wayne State University.
Capturing_the_Potential_of_Cloud_1_5MB.pdf
(Accessed on 02.08.2012), http://wirelesshealth.wayne.edu/
10. Intel Science and Technology for Cloud Computing.
(Accessed on 05.08.2012), http://www.cmu.edu/homepage/
society/2009/winter/a-break-in-the-clouds.shtml
http://www.istc-cc.cmu.edu/index.shtml

235 International Journal of Information Dissemination and Technology | October-December 2013 | Vol. 3 | Issue 4