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

net/publication/228719828

A Survey on Open-source Cloud Computing Solutions

Article · January 2010

CITATIONS READS

87 4,237

5 authors, including:

Patricia Takako Endo Glauco Estacio Gonçalves

Universidade de Pernambuco Federal University of Pará
204 PUBLICATIONS 1,869 CITATIONS 79 PUBLICATIONS 793 CITATIONS

SEE PROFILE SEE PROFILE

Judith Kelner Djamel Sadok

Federal University of Pernambuco Federal University of Pernambuco
327 PUBLICATIONS 2,764 CITATIONS 334 PUBLICATIONS 2,928 CITATIONS

SEE PROFILE SEE PROFILE

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

IMPReSS - Intelligent System Development Platform for Intelligent and Sustainable Society View project

ISOLA AI View project

All content following this page was uploaded by Glauco Estacio Gonçalves on 20 May 2014.

The user has requested enhancement of the downloaded file.

 VIII Workshop em Clouds, Grids e Aplicações 3

A Survey on Open-source Cloud Computing Solutions

Patrícia Takako Endo1, Glauco Estácio Gonçalves1, Judith Kelner1, Djamel Sadok1
1
Universidade Federal de Pernambuco – UFPE
Caixa Postal 7851 – 50732-970 – Recife – PE
{patricia,glauco,jk,jamel}@gprt.ufpe.br

Abstract. Cloud computing is an attractive computing model since it allows

for resources to be provisioned according on a demand basis, i.e., cloud users
can rent resources as they become necessary. This model motivated several
academic and non-academic institutions to develop open-source cloud
solutions. This paper presents and discusses the state-of-the of open-source
solutions for cloud computing. The authors hope that the observation and
classification of such solutions can leverage the cloud computing research
area providing a good starting point to cope with some of the problems
present in cloud computing environments.

1. Introduction
At present, it is common to access content across the Internet with little reference to the
underlying hosting infrastructure, which is formed by data centers maintained by content
providers. The entire technology used to provide such level of locality transparency
offers also a new model for the provision of computing services, stated as Cloud
Computing. In such a model, enterprises put their applications in the cloud – a very
large computational entity from a developer´s perspective – without concern for where
the services are actually hosted or how they are delivered. Through access to slice of
computational power over a scalable network of nodes, enterprises can reduce or
eliminate costs associated with internal infrastructure for the provision of their services.
A further viewpoint of cloud users, in addition to costs reduction, comes from
the fact that the cloud computing model can be attractive since it allows resources to be
provisioned according to the enterprise needs. This is in contrast, to traditional practices,
where resources are often dimensioned according to the worst case use and peak
scenarios. Hence cloud users can rent resources as they become necessary, in a much
more scalable and elastic way. Moreover, enterprises can transfer operational risks to
cloud providers. In the viewpoint of cloud providers, the model offers a way for better
utilization of their own infrastructure. Authors in [Armbrust et al 2009] point that this
model benefits from a form of statistical multiplexing, since it allocates resources for
several users concurrently. This statistical multiplexing of data centers is guaranteed
through several decades of research in many areas as: distributed computing, grid
computing, web technologies, service computing, and virtualization. Several authors
([Armbrust et al 2009], [Vaquero et al 2008], [Buyya et al 2009]) agree that from such
research areas virtualization brought the key technologies to leverage cloud computing.
Despite the increasingly widespread use of the Cloud Computing term, there is
no formal definition for it yet. In a recent paper [Vaquero et al 2008], authors review the
4 Anais

cloud literature for a minimum set of characteristics that cloud solutions must present.
But they were not able to find even a single common feature in literature definitions.
They note that the set of characteristics that at most are similar to a minimum common
definition are: scalability, pay-per use model, and virtualization. Finally, using these
features the authors gave their own definition: “Clouds are a large pool of easily usable
and accessible virtualized resources. These resources can be dynamically reconfigured to
adjust to a variable load, allowing also for an optimum resource utilization. This pool of
resources is typically exploited by a pay-per-use model in which guarantees are offered
by the Infrastructure Provider by means of customized SLA”.
Since the popularization of the cloud computing term in 2007, with IBM Blue
Cloud [Vouk 2008], several enterprises become cloud computing providers: Amazon
and their Elastic Compute Cloud (EC2)1, Google with Google App Engine2, Microsoft
with their Windows Azure Platform3, Salesforce and their Force.com4 and so on.
Though these solutions fit the cloud computing definition they differ in their
programmability. A concept borrowed from the network virtualization area [Chowdhury
and Boutaba 2010], programmability is relative to the programming features a network
element offers to developers, measuring how much freedom the developer has to
manipulate resources and/or devices.
This concept can be applied to compare cloud computing solutions. More
programmable clouds offer environments where developers are free to choose their own
programming paradigm, languages, and platforms, having total control over their
virtualized resources. Less programmable clouds restrict developers in some way:
forcing a set of programming languages, or allowing support for only one application
paradigm. On the other hand, a high level of programmability turns hard the cloud
management because the cloud provider will have a much more heterogeneous
environment to manage. For example, if a cloud provider allows their users to use any
operating system in a virtual machine, the cloud operator will have to cope with a large
number of solutions to provide fault-tolerance services. Moreover, less programmable
solutions abstract some operational issues (processes communication, network access,
storage functions, etc.) through some type of middleware. An instance of a cloud
solution with high programmability is the Amazon EC2, where users can choose their
operating system from a set of supported operating systems given by Amazon and they
can configure their virtual machines to work as they see fit. The Google App Engine is
an example of a less programmable solution, since it restricts developers to the Web
paradigm and to some programming languages.
A common characteristic present in all the cited cloud solutions is that cloud
owners understandably avoid revealing the underlying features of their solution, since
this is seen as part of their strategic information. Despite this, one may point out the
considerable efforts made by several academic and non-academic institutions to develop
open-source cloud solutions. The authors hope that the observation and classification of

1
http://aws.amazon.com/ec2/
2
http://code.google.com/intl/appengine/appengine/
3
http://www.microsoft.com/windowsazure/
4
http://www.salesforce.com/platform/
 VIII Workshop em Clouds, Grids e Aplicações 5

such solutions can leverage further the cloud computing research area providing a good
starting point to discovering different approaches to the problems present in cloud
computing environments.
This paper presents and discusses the state-of-the of open-source solutions for
cloud computing. The remainder of this paper is organized as follows. Next section
introduces the main challenges developers face to create cloud computing solutions.
Section III shows efforts in standardizing cloud computing interfaces for
interoperability. Section IV introduces the main open-source cloud architectures and
Section V makes comparisons between the architectures. Finally, a conclusion of this
paper and future works are presented in Sections VI.

2. Challenges
The development of cloud computing solutions brings several technical challenges to
cloud developers. These challenges can be grouped in three main areas: negotiation,
decision, and operation. In the negotiation area, these are the challenges relative to how
application developers interface with the cloud as well as the description of the cloud
offerings. It includes also the definition of the programmability level that the cloud
solution will offer. The decision area copes with the main problem that clouds faces
behind the scenes: How virtual resources can be scheduled to meet user requirements,
for example? Last, the operation area is associated with the enforcement of decisions
and the communication between cloud elements. The following sub-sections discuss in
details the main challenges in each one of these areas.

2.1. Negotiation
The negotiation area concerns itself with challenges relative to the interface between the
application developers and the cloud. Generally, the interface between the cloud and
application developers assumes the form of an Application Programming Interface
(API), but, depending on the programmability level offered by the cloud, this API can be
implemented in several ways ranging from a web-service based toolkit to control virtual
machines in the cloud to a set of programming primitives used to develop distributed
applications in the cloud. In addition to these basic functions, such APIs must allow
developers to request – and control, possibly – additional functionalities offered by a
cloud operator like service quality assessment, load balance, elastic application growth,
backup strategies, and so on. There are still some other requirements that can be
considered in the cloud API like geographical restrictions enforced to allocate virtual
machines due to legal issues. One may think of some type of content or application that
is strategically limited to a country or a region for copyright or security reasons.
At the present, APIs and the negotiation process offered by cloud providers
follow a semi-automatic scheme where a human interacts with the cloud through
programming primitives or a visual interface. But, next generation clouds can offer
sophisticated ways to interact with human users through high-level abstractions and
service-level policies. Such an interface type will need some formalism to specify both
cloud offerings and application requirements as well as offering the support for an
automatic negotiation process.
6 Anais

2.2. Decision
The main target of any cloud operator is to schedule developer applications aiming for
the maximum utilization of cloud resources. A developer´s application covers, beyond
the actual code, some additional information about application´s needs and services
negotiated previously. In other words, one can abstract this additional information to
some type of network virtualization demand with a topology formed by virtual nodes
where the application runs and virtual links for communication. Thus, the cloud
operator problem turns into that of selecting the best suitable physical resources to
accommodate these virtual resources.
This careful mapping requires advanced strategies. The first challenge imposed
by this optimization problem is that it is NP-hard [Andersen 2002] and hence any useful
solution would need to relax some of its problem conditions and constraints to obtain an
approximate solution in a polynomial time. The second challenge is to meet all the
clauses negotiated with the developer. Depending on the nature of such contract,
application scheduling algorithms will cope with quality restrictions, jurisdiction
restrictions, elastic growth, and so on.

2.3. Operation
Metaphorically, one can say that while in the decision area the cloud operator must
identify solutions for the “brain” of the cloud, in the operation area it must attack the
problems of the “limbs” of the cloud, i.e., they must provide some form to enforce
decisions. The enforcement here covers the communication protocols and the
configuration of cloud elements.
A communication protocol can be used to monitor and reserve resources in the
cloud. The cloud is composed by different elements like processing servers, switches,
routers, links and storage components. Due to such heterogeneity, the communication
between the decision-maker and elements puts a challenge on cloud design. Overall,
existing cloud solutions use Web Services to provide communication with processing
and storage nodes, but many communication elements do not support such
implementations. Thus, cloud architects are using traditional traffic engineering
protocols to provide reservation of network elements. One possible idea to cope with
this challenge is to use smart communication nodes with an open interface to create new
services in the node the emerging Openflow-enabled switches [McKeown et al 2008].
Node communication is just one part of the problem; the other one is to
configure this. Here, the recent advances in server virtualization have provided several
solutions for operators to benefit from.

3. Standardization efforts
A considerable challenge present in many of the raised discussions around the cloud is
related to the need for standardization. All the three areas presented in Section 2 face the
standardization challenge in some way, but the main challenge occurs in the negotiation
area. Currently, cloud providers offer proprietary interfaces to access their services. This
locks users within a given provider as they cannot migrate their applications and
services easily between cloud providers [Buyya et al 2009]. It is hoped that cloud
 VIII Workshop em Clouds, Grids e Aplicações 7

providers see such a problem and work together to offer a standardized API based on
open standards like SOAP and REST.
An important effort in the standardization comes from the Open Cloud
Manifesto [OpenCloud 2009]. This is an initiative supported by hundreds of companies
that aims to discuss with cloud organizations a way to produce open standards for cloud
computing. Their major doctrines are collaboration and coordination of efforts on the
standardization, adoption of open standards wherever appropriate, and the development
of standards based on customer requirements. Participants of the Open Cloud Manifesto
through the Cloud Computing Use Case group produced an interesting white paper
[OpenCloud 2010] highlighting the requirements that need to be standardized in a cloud
environment to ensure interoperability in the most typical scenarios of interaction – Use
Cases – in cloud computing.

4. Open-source solutions for Cloud Computing

Due to the large growth of cloud computing, there are several solutions in this area. This
article is focused on open source solutions, highlighting their main characteristics and
architectures proposed.

4.1. Xen Cloud Platform (XCP)

The Xen hypervisor [Citrix Systems 2010b] is a solution for infrastructure virtualization
that provides an abstraction layer between servers’ hardware and the operating system.
A Xen hypervisor allows each physical server to run several “virtual servers” handling
the operating system and its applications from the underlying physical server. The Xen
solution is used by many cloud solutions such as Amazon EC2, Nimbus and Eucalyptus.
Recently, Xen.org announced the Xen Cloud Platform (XCP) [Citrix Systems
2010a] as a solution for cloud infrastructure virtualization. But, differently from existent
open source cloud solutions, XCP does not provide the overall architecture for cloud
services. Their goal is to provide a tool to cope with automatic configuration and
maintenance of cloud platforms [Citrix Systems 2010c].
XCP Resource Pool

Private
Network
Master XCP Host Shared Storage

XCP Host XCP Host XCP Host XCP Host XCP Host

Figure 1. XCP Architecture

The XCP architecture (Figure 1) is based on the XCP hosts that are responsible
to host the virtual machines. According to [Xen.org. 2009], these hosts are aggregated in
8 Anais

a XCP resource pool and using a Shared Storage the virtual machines can be started
and restarted on any XCP host. The Master XCP host offers an administration interface
and forwards command messages to others XCP hosts.

4.2. Nimbus
Nimbus [Keahey 2009] is an open source solution (licensed under the terms of the
Apache License) to turn clusters into an Infrastructure as a Service (IaaS) for Cloud
Computing focusing mainly on scientific applications.
This solution gives to users the possibility to allocate and configure remote
resources by deploying VMs – known as Virtual Workspace Service (VWS). A VWS is
a VM manager that different frontends can invoke.
To deploy applications, Nimbus offers a “cloudkit” configuration that consists of
a manager service hosting and an image repository. The workspace components are
shown in Figure 2.

Figure 2. Nimbus workspace components [Keahey 2009]

• Workspace service: is web services based and provides security with the GSI
authentication and authorization. Currently, Nimbus supports two frontends:
Amazon EC2 and WSRF.
• Workspace control: is responsible for controlling VM instances, managing and
reconstructing images, integrating a VM to the network and assigning IP and
MAC addresses. The workspace control tools operate with the Xen hypervisor
and can also operate with KVM5.
• Workspace resource management: is an open source solution to manage
different VMs, but can be replaced by other technologies such as OpenNebula.
• Workspace pilot: is responsible for providing virtualization with few changes in
cluster operation. This component handles signals and has administration tools.

4.3. OpenNebula
OpenNebula [OpenNebula Project 2010] is an open-source toolkit used to build private,
public and hybrid clouds. It has been designed to be integrated with networking and
storage solutions and to fit into existing data centers.

5
http://www.linux-kvm.org/page/Main_Page
 VIII Workshop em Clouds, Grids e Aplicações 9

The OpenNebula architecture (Figure 3) is based on three basic technologies to

enable the provision of services on a distributed infrastructure: virtualization, storage
and network. All resource allocation is done based on policies.

Figure 3. OpenNebula architecture [OpenNebula Project 2010]

The Cumulus Project [Wang et al 2008] is an academic proposal based on

OpenNebula. Cumulus intends to provide virtual machines, virtual applications and
virtual computing platforms for scientific applications. Visualizing the integration of
already existing technologies, the Cumulus project uses HP and IBM blade serves
running Linux and Xen hypervisor.
The Cumulus networking solution was called the “forward” mode, where users
do not need to specify any network configuration information. Instead the backend
servers are responsible for allocating a dynamic IP address for a VM and returning these
to the users, making such networking solution transparent to the users.
The Cumulus design is a layered architecture (Figure 4) with three main entities:
Cumulus frontend, OpenNebula frontend, and OS Farm. This proposal focuses on
reaching scalability and autonomy of data centers.

Figure 4. Cumulus architecture [Wang et al 2008]

10 Anais

• Cumulus frontend: the Cumulus frontend is the access point for a Cumulus
system and is responsible for handling VM requirements.
• OpenNebula frontend: the OpenNebula frontend provides an interface to
manage the distributed blade servers and the resources for VM deployment. To
administrate a common user system, Cumulus uses NIS (Network Information
System) and NFS (Network File System) to manage shared directory. Moreover,
OpenNebula was merged with secure infrastructure solutions, such as LDAP
(Lightweight Directory Access Protocol) and the Oracle Cluster File System.
• OS Farm: the OS Farm is a tool for VM template management that operates to
generate and to store Xen VM images and virtual appliances.

4.4. Eucalyptus
Eucalyptus [Nurmi et al 2009] is an open source cloud computing framework focused
on academic research. It provides resources for experimental instrumentation and study.
Eucalyptus users are able to start, control, access and terminate entire virtual machines.
In its current version, Eucalyptus supports VMs that run atop the Xen supervisor
[Barham et al 2003].
According to [Nurmi et al 2009], the Eucalyptus project presents four
characteristics that differentiate it from others cloud computing solutions: a) Eucalyptus
was designed to be simple without requiring dedicated resources; b) Eucalyptus was
designed to encourage third-party extensions through modular software framework and
language-agnostic communication mechanisms; c) Eucalyptus external interface is
based on the Amazon API (Amazon EC2) and d) Eucalyptus provides a virtual network
overlay that both isolates network traffic of different users and allows clusters to appear
to be part of the same local network.
The Eucalyptus architecture is hierarchical (Figure 5) and made up of four high
level components, where each one is implemented as a stand-alone web service.
• Node Controller (NC): this component runs on every node that is destined for
hosting VM instances. An NC is responsible to query and control the system
software (operating system and hypervisor) and for conforming requests from its
respective Cluster Controller. The role of NC queries is to collect essential
information, such as the node’s physical resources (e.g. the number of cores and
the available disk space) and the state of VM instances on the nodes. NC sends
this information to its Cluster Controller (CC). NC is also responsible for
assisting CC to control VM instances on a node, verifying the authorization,
confirming resources availability and executing the request with the hypervisor.
• Cluster Controller (CC): this component generally executes on a cluster front-
end machine, or any machine that has network connectivity to two nodes: one
running NCs and another running the Cloud Controller (CLC). A CC is
responsible to collect/report information about and schedule VM execution on
specific NCs and to manage virtual instance network overlay.
• Storage Controller (Walrus): this component is a data storage service that
provides a mechanism for storing and accessing virtual machine images and user
 VIII Workshop em Clouds, Grids e Aplicações 11

data. Walrus is based on web services technologies and compatible with

Amazon’s Simple Storage Service (S3) interface [Amazon 2006].
• Cloud Controller (CLC): this component is the entry-point into the cloud for
users. Its main goal is to offer and manage the Eucalyptus underlying virtualized
resources. CLC is responsible for querying node managers for resources’
information, making scheduling decisions, and implementing them by requests
to CC. This component is composed by a set of web services which can be
grouped into three categories, according their roles: resource services, data
services, and interface services.
CLC and Walrus

Public
CC Network CC

Private Private
Network Network
NC

NC

NC

NC

NC

NC

NC

NC
Cluster A Cluster B

Figure 5. Eucalyptus architecture [Nurmi et al 2009]

Ubuntu Enterprise Cloud (UEC)6 is an Amazon EC2 like infrastructure and is

powered by Eucalyptus. Its main goal is to provide a simple process of building and
managing internal infrastructure for cloud. The Ubuntu 9.04 Server Edition is integrated
with Eucalyptus that uses the KVM hypervisor.
The UEC architecture is based on the Eucalyptus architecture in which each
elements is an independent web service that publishes a Web Service Description
Language (WSDL) document defining the API to interact with it.
Furthermore, UEC defines three layers for security: authentication and
authorization, network isolation and Machine Instance Isolation (MInst). The
authentication and authorization layer is responsible for locally generated X.509
certificates; the network isolation layer is important to prevent eavesdropping of
network traffic and; the MInst layer consists of Networking isolation, Operating System
isolation, and Hypervisor based machine isolation.

4.5. TPlatform
TPlatform [Peng et al 2009] is a cloud solution that provides a development platform for
web mining applications, which is inspired in Google cloud technologies, and which
acts as a Platform as a Service (PaaS) solution. Their infrastructure is supported by three
technologies: a scalable file system called Tianwang File System (TFS) what is similar
to the Google File System (GFS), the BigTable data storage mechanism, and the

6
http://www.ubuntu.com/cloud
12 Anais

MapReduce programming model. The TPlatform framework is composed by three

layers (Figure 6):

Figure 6. TPlatform framework [Peng et al 2009]

• PC Cluster: this layer provides the hardware infrastructure for data processing.
• Infrastructure: this layer consists of file system (TFS), distributed data storage
mechanism (BigTable), and programming model (MapReduce).
• Data Processing Applications: this layer provides the services for users to
develop their application (e.g. web data analysis and language processing).

4.6. Apache Virtual Computing Lab (VCL)

Apache VCL [VCL 2010] is an open-source solution for the remote access over the
Internet to dynamically provision and reserve computational resources for diverse
applications, acting as Software as a Service (SaaS) solution. VCL has a simple
architecture formed by three tiers:
• Web server: represents the VCL portal and uses Linux/Apache/PHP solution.
This portal provides an user interface that enable the requesting and
management of VCL resources;
• Database server: storages information about VCL reservations, access controls,
machine and environment inventory. It uses Linux/SQL solution;
• Management nodes: is the processing engine. A management node controls a
subset of VCL resources, which may be physical blade servers, traditional rack,
or virtual machines. It uses Linux/VCLD (perl)/image library solution. VCLD is
a middleware responsible to process reservations or jobs assigned by the VCL
web portal. According to type of environment requested, VCLD should assure
that service (computational environment) will be available to user.
Figure 7 shows a conceptual overview of the VCL, where the user must connect
firstly to the VCL Scheduling Application in order to access its resources through a web
interface. Users may request a reservation to use the environment immediately or
schedule to use it in the future.
 VIII Workshop em Clouds, Grids e Aplicações 13

Figure 7. Apache conceptual overview [VCL 2010]

4.7. Enomaly Elastic Computing Platform

Enomaly ECP Community Edition under the AGPL license [Enomaly 2009] is the open
source cloud solution offered by Enomaly Inc. This version focuses on virtual machine
administration in small clouds environments. Compared with the Enomaly commercial
solution (called Service Provider Edition), the Enomaly open source edition suffers from
many restrictions, such as limited scalability, no capacity control mechanism, no support
for accounting and metering, and so on.

5. Discussion
As pointed out earlier in this paper, there are several solutions for cloud computing7
focusing on different areas and ranging from hardware resource outsourcing to user
services providing. Each solution presents a different vision about cloud architecture
and implementation. Moreover, each approach has an implication that directly impacts
its business model: the closer to the hardware level, the more options a user can handle
but at the cost of having to configure her cloud (more configuration flexibility).
Amazon EC2 and IBM Capacity on Demand (CoD) are solutions that offer to
their users this configuration flexibility. In this business model, users can choose and
configure computational resources at the hardware level and OS levels. At the other
extreme, solutions like Google App Engine and Windows Azure, try to turn
development easy to their users, but at the same time, confine them to specific APIs and
software platforms. Moreover, solutions like JoliCloud8 are more limited as they offer a
single service (operating system). In the middle, there are solutions that offer a
middleware-like approach to users, where the hardware resources can be configured and
handled subject to some restrictions and where applications can also be developed.
All the presented open-source solutions and the cited commercial solutions are
categorized into Figure 8. The graphic compares solutions and their business model
(hardware, middleware and user level) according to configuration flexibility.

7
More than 500 Cloud solutions have been reported.
8
http://www.jolicloud.com/
14 Anais

configuration
flexibility

XCP
Eucalyptus
Amazon EC2
IBM CoD
Nimbus
OpenNebula
Cumulus
Apache VCL
Enomaly ECP
Google App Engine
Windows Azure
UEC
JoliCloud

hardware user level user

level middleware level

Figure 8. Cloud computing solutions

Finally, Table 1 presents a comparative board of the open source cloud solutions
described in this paper, in terms of the service type (IaaS, PaaS, and SaaS), the main
characteristics, and the infrastructure technologies. The table also cites some users of
each cloud solution.
Table 1. Comparison between open-source Cloud Computing solutions
Solutions Service Main Characteristics Infrastructure Used by
XCP IaaS Only a tool for automatic Xen XCP community
maintenance of clouds
Nimbus IaaS Aims to turn legacy Xen hypervisor Brookhaven
clusters into IaaS Clouds and KVM National Labs9
OpenNebula IaaS Policy-driven resource Xen hypervisor Cumulus Project
allocation
Eucalyptus IaaS Hierarchical Xen hypervisor UEC
Architecture and KVM
TPlatform PaaS Focus on web text TFS, BigTable TPlatform Project10
mining applications and MapReduce
Apache VCL SaaS Internet access for VMware Educational11 and
several applications Government12 users
Enomaly IaaS Open version is focused Xen, KVM and Several companies13
in small clouds VirtualBox

9
http://www.bnl.gov/rhic/default.asp
10
http://net.pku.edu.cn/~webg/tplatform/
11
East Carolina University, Johnston Community College, North Carolina Central University, University
of North Carolina at Greensboro, Wake Technical Community College and Western Carolina University
12
North Carolina Community College System
13
Some companies that use Enomaly Service Provider Edition: Orange/France Télécom, Bank of China,
City Network, CentriLogic
 VIII Workshop em Clouds, Grids e Aplicações 15

6. Conclusions and Future Works

There is a clear need for the standardization of current cloud platforms at least of terms
of interface, negotiation and access through Web services. Understandably, this is a
considerable task as many clouds use different abstraction levels, some are generic
whereas others focus on a specific application domain, etc. Some initial steps have been
taken into this direction with the setup of the Open Cloud Manifesto, an initiative
supported by hundreds of companies. In the mean time we will continue to see some
clouds such as Nimbus implementing a number of front ends (e.g. Amazon EC2 and
WSRF) to ensure access to their existing users.
Interestingly, some solutions such as the OpenNebula have been first to adopt
policies for resource management. The use of policies remains a challenge in many
areas and clouds may benefit from it. It is also important to acknowledge the leadership
and string presence of academic efforts such as Eucalyptus and Xen. These have been at
the forefront of inovation supported by the many commercial cloud systems that
currently are based on these. These efforts are expected to continue as more work is
needed to remove much of the misticism and conflicts surrounding the use of clouds.
As future work, authors proposes a quantitative comparison of the presented
solutions through performance evaluation measurements.

References
Amazon. (2006) “Amazon simple storage service (Amazon S3) – API Reference”,
http://docs.amazonwebservices.com/AmazonS3/2006-03-01/.
Andersen, D. (2002) “Theoretical Approaches to Node Assignment”, Unpublished
manuscript. http://www.cs.cmu.edu/dga/papers/andersen-assign.ps
Armbrust, M., Fox, A., Griffith, R., Joseph, A.D., Katz, R.H., Konwinski, A., Lee, G.,
Patterson, D.A., Rabkin, A., Stoica, I., Zaharia, M. (2009) “Above the Clouds: A
Berkeley View of Cloud Computing”, Tech. Rep. UCB/EECS-2009-28, EECS
Department, University of California, Berkeley.
Barham, P., Dragovic, B., Fraser, K., Hand, S., Harris, T., Ho, A., Neugebauer, R., Pratt,
I. and Warfield, A. (2003) “Xen and the art of virtualization”. In: 19th ACM
symposium on Operating systems principles, New York, NY, USA.
Buyya, R., Yeo, C.S., Venugopal, S., Broberg, J., Brandic, I. (2009) “Cloud computing
and emerging IT platforms: Vision, hype, and reality for delivering computing as the
5th utility”. In: Future Generation Computer Systems, Elsevier B. V.
Chowdhury, N.M. M. K. and Boutaba, R. (2010) “A survey of network virtualization”.
In: Computer Networks, volume 54, issue 5, pages 862-876. Elsevier B. V.
Citrix Systems. (2010) “Xen Cloud Platform - Advanced Virtualization Infrastructure
for the Clouds”, http://www.xen.org/products/cloudxen.html.
Citrix Systems. (2010) “Xen Hypervisor”, http://xen.org/products/xenhyp.html.
Citrix Systems. (2010) “The Xen Cloud Project”, The Citrix Blog. Available at
http://community.citrix.com/pages/viewpage.action?pageId=81690805
 16 Anais

Enomaly. (2009) “Enomaly – Elastic Computing”. http://src.enomaly.com/.

Keahey, K. (2009) “Nimbus: Open Source Infrastructure-as-a-Service Cloud Computing
Software”, Workshop on adapting applications and computing services to multi-core
and virtualization, CERN, Switzerland.
McKeown, N., Anderson, T., Balakrishnan, H., Parulkar, G., Peterson, L., Rexford, J.,
Shenker, S. and Turner, J. (2008) “OpenFlow: enabling innovation in campus
networks”, ACM SIGCOMM Computer Communication Review.
Nurmi, D., Wolski, R., Grzegorczyk, C., Obertelli, G., Soman, S., Youseff, L. and
Zagorodnov, D. (2009) “The Eucalyptus Open-Source Cloud Computing System”.
In: 9th IEEE/ACM International Symposium on Cluster Computing and the Grid.
OpenCloud. (2009) “The Open Could Manifesto”. http://www.opencloudmanifesto.org/.
OpenCloud. (2010) “Cloud Computing Use Cases Whitepaper - Version 3.0”.
http://groups.google.com/group/cloud-computing-use-cases?hl=en.
OpenNebula Project. (2010) “OpenNebula.org – The Open Source Toolkit for Cloud
Computing”, http://www.opennebula.org/.
Peng, B., Cui, B. and Li, X. (2009) “Implementation Issues of A Cloud Computing
Platform”. IEEE Data Engineering Bulletin, volume 32, issue 1.
Vaquero, L.M., Merino, L. R. Caceres, J. Lindner, M. (2008) “A break in the clouds:
towards a cloud definition”. ACM SIGCOMM Computer Communication Review.
VCL. (2010). “Virtual Computing Lab”. https://cwiki.apache.org/VCL/.
Vouk, M.A. (2008) “Cloud Computing – Issues, Research and Implementations”. In:
Journal of Computing and Information Technology. University of Zagreb, Croatia.
Xen.org. (2009) “Xen Cloud Platform Administrator’s Guide: Release 0.1”.
Wang, L., Tao, J., Kunze, M., Rattu, D. and Castellanos, A. (2008) “The Cumulus
Project: Build a Scientific Cloud for a Data Center”. In: Workshop on Cloud
Computing and Its Applications, Chicago, USA.

View publication stats