Community Cloud Computing.

Full text

Community Cloud Computing
Alexandros Marinos
Department of Computing
University of Surrey
United Kingdom
e-mail: a.marinos@surrey.ac.uk
Gerard Briscoe
Department of Media and Communications
London School of Economics and Political Science
United Kingdom
e-mail: g.briscoe@lse.ac.uk
Abstract—.Cloud Computing is rising fast, with its data
centres growing at an unprecedented rate. However, this has
come with concerns over privacy, efficiency at the expense
of resilience, and environmental sustainability, because of
the dependence on Cloud vendors such as Google, Amazon
and Microsoft. Our response is an alternative model for the
Cloud conceptualisation, providing a paradigm for Clouds in
the community, utilising networked personal computers for
liberation from the centralised vendor model. Community
Cloud Computing (C3) offers an alternative architecture,
created by combing the Cloud with paradigms from Grid
Computing, principles from Digital Ecosystems, and sus-
tainability from Green Computing, while remaining true to
the original vision of the Internet. It is more technically
challenging than Cloud Computing, having to deal with
distributed computing issues, including heterogeneous nodes,
varying quality of service, and additional security constraints.
However, these are not insurmountable challenges, and with
the need to retain control over our digital lives and the
potential environmental consequences, it is a challenge we
must pursue.
Index Terms—.Cloud Computing, Community Cloud,
Community Cloud Computing, Green Computing, Sustain-
ability.
I. INTRODUCTION
The recent development of Cloud Computing provides
a compelling value proposition for organisations to out-
source their Information and Communications Technology
(ICT) infrastructure [1]. However, there are growing con-
cerns over the control ceded to large Cloud vendors [2],
especially the lack of information privacy [3]. Also, the
data centres required for Cloud Computing are growing
exponentially [4], creating an ever-increasing carbon foot-
print and therefore raising environmental concerns [5], [6].
The distributed resource provision from Grid Com-
puting, distributed control from Digital Ecosystems, and
sustainability from Green Computing, can remedy these
concerns. So, Cloud Computing combined with these
approaches would provide a compelling socio-technical
conceptualisation for sustainable distributed computing,
utilising the spare resources of networked personal com-
puters collectively to provide the facilities of a virtual
data centre and form a Community Cloud. Therefore,
essentially reformulating the Internet to reflect its current
uses and scale, while maintaining the original intentions
[7] for sustainability in the face of adversity. Including
extra capabilities embedded into the infrastructure which
would become as fundamental and invisible as moving
packets is today.
II. CL OU D COMPUTING
Cloud Computing is the use of Internet-based technolo-
gies for the provision of services [1], originating from the
cloud as a metaphor for the Internet, based on depictions in
computer network diagrams to abstract the complex infras-
tructure it conceals [8]. It can also be seen as a commercial
evolution of the academic-oriented Grid Computing [9],
succeeding where Utility Computing struggled [10], [11],
while making greater use of the self-management advances
of Autonomic Computing [12]. It offers the illusion of
infinite computing resources available on demand, with
the elimination of upfront commitment from users, and
payment for the use of computing resources on a short-
term basis as needed [3]. Furthermore, it does not require
the node providing a service to be present once its service
is deployed [3]. It is being promoted as the cutting-edge
of scalable web application development [3], in which
dynamically scalable and often virtualised resources are
provided as a service over the Internet [13], [1], [14],
[15], with users having no knowledge of, expertise in, or
control over the technology infrastructure of the Cloud
supporting them [16]. It currently has significant momen-
tum in two extremes of the web development industry [3],
[1]: the consumer web technology incumbents who have
resource surpluses in their vast data centres1, and various
consumers and start-ups that do not have access to such
computational resources. Cloud Computing conceptually
incorporates Software-as-a-Service (SaaS) [18], Web 2.0
[19] and other technologies with reliance on the Internet,
providing common business applications online through
web browsers to satisfy the computing needs of users,
while the software and data are stored on the servers.
Figure 1 shows the typical configuration of Cloud Com-
puting at run-time when consumers visit an application
served by the central Cloud, which is housed in one
or more data centres [20]. Green symbolises resource
consumption, and yellow resource provision. The role of
coordinator for resource provision is designated by red,
and is centrally controlled. Even if the central node is
implemented as a distributed grid, which is the usual
incarnation of a data centre, control is still centralised.
Providers, who are the controllers, are usually companies
with other web activities that require large computing
1Adata centre is a facility, with the necessary security devices and
environmental systems (e.g. air conditioning and fire suppression),
for housing a server farm, a collection of computer servers that can
accomplish server needs far beyond the capability of one machine
[17].
arXiv:0907.2485v3 [cs.NI] 12 Oct 2009

Figure 1. Cloud Computing: Typical configuration when consumers
visit an application served by the central Cloud, which is housed in one
or more data centres [20]. Green symbolises resource consumption, and
yellow resource provision. The role of coordinator for resource provision
is designated by red, and is centrally controlled.
resources, and in their efforts to scale their primary busi-
nesses have gained considerable expertise and hardware.
For them, Cloud Computing is a way to resell these
as a new product while expanding into a new market.
Consumers include everyday users, Small and Medium
sized Enterprises (SMEs), and ambitious start-ups whose
innovation potentially threatens the incumbent providers.
A. Layers of Abstraction
While there is a significant buzz around Cloud Comput-
ing, there is little clarity over which offerings qualify or
their interrelation. The key to resolving this confusion is
the realisation that the various offerings fall into different
levels of abstraction, as shown in Figure 2, aimed at
different market segments.
1) Infrastructure-as-a-Service (IaaS) [21]: At the most
basic level of Cloud Computing offerings, there are
providers such as Amazon [22] and Mosso [23], who
provide machine instances to developers. These instances
essentially behave like dedicated servers that are controlled
by the developers, who therefore have full responsibility
for their operation. So, once a machine reaches its perfor-
mance limits, the developers have to manually instantiate
another machine and scale their application out to it.
This service is intended for developers who can write
arbitrary software on top of the infrastructure with only
small compromises in their development methodology.
2) Platform-as-a-Service (PaaS) [24]: One level of ab-
straction above, services like Google App Engine [25] pro-
vide a programming environment that abstracts machine
instances and other technical details from developers. The
programs are executed over data centres, not concerning
the developers with matters of allocation. In exchange for
this, the developers have to handle some constraints that
the environment imposes on their application design, for
example the use of key-value stores2instead of relational
databases.
3) Software-as-a-Service (SaaS) [18]: At the
consumer-facing level are the most popular examples of
Cloud Computing, with well-defined applications offering
2A distributed storage system for structured data that focuses on
scalability, at the expense of the other benefits of relational databases
[26], e.g. Google’s BigTable [27] and Amazon’s SimpleDB [28].
PaaS
(Platform as a Service)
IaaS
(Infrastructure as a Service)
SaaS
(Software as a Service)
Vendor Developers
End Users
Provide
Provide
Provide
Consume
Consume
Provide
Support
Support
Consume
Figure 2. Abstractions of Cloud Computing: While there is a significant
buzz around Cloud Computing, there is little clarity over which offerings
qualify or their interrelation. The key to resolving this confusion is
the realisation that the various offerings fall into different levels of
abstraction, aimed at different market segments.
users online resources and storage. This differentiates
SaaS from traditional websites or web applications which
do not interface with user information (e.g. documents)
or do so in a limited manner. Popular examples include
Microsoft’s (Windows Live) Hotmail, office suites such
as Google Docs and Zoho, and online business software
such as Salesforce.com.
To better understand Cloud Computing we can cat-
egorise the roles of the various actors. The vendor as
resource provider has already been discussed. The ap-
plication developers utilise the resources provided, build-
ing services for the end users. This separation of roles
helps define the stakeholders and their differing interests.
However, actors can take on multiple roles, with vendors
also developing services for the end users, or developers
utilising the services of others to build their own services.
Yet, within each Cloud the role of provider, and therefore
controller, can only be occupied by the vendor providing
the Cloud.
B. Concerns
The Cloud Computing model is not without concerns, as
others have noted [29], [3], and we consider the following
as primary:
1) Failure of Monocultures: The uptime3of Cloud
Computing based solutions is an advantage, when com-
pared to businesses running their own infrastructure, but
often overlooked is the co-occurrence of downtime in
vendor-driven monocultures. The use of globally decen-
tralised data centres for vendor Clouds minimises failure,
aiding its adoption. However, when a cloud fails, there
is a cascade effect crippling all organisations dependent
on that Cloud, and all those dependent upon them. This
was illustrated by the Amazon (S3) Cloud outage [31],
which disabled several other dependent businesses. So,
failures are now system-wide, instead of being partial or
3Uptime is a measure of the time a computer system has been running,
i.e. up. It came into use to describe the opposite of downtime, times
when a system was not operational [30].

localised. Therefore, the efficiencies gained from central-
ising infrastructure for Cloud Computing are increasingly
at the expense of the Internet’s resilience.
2) Convenience vs Control: The growing popularity of
Cloud Computing comes from its convenience, but also
brings vendor control, an issue of ever-increasing concern.
For example, Google Apps for in-house e-mail typically
provides higher uptime [32], but its failure [33] highlights
the issue of lock-in that comes from depending on vendor
Clouds. The even greater concern is the loss of information
privacy, with vendors having full access to the resources
stored on their Clouds. So much so the British government
is considering a ‘G Cloud’ for government business appli-
cations [34]. In particularly sensitive cases of SMEs and
start-ups, the provider-consumer relationship that Cloud
Computing fosters between the owners of resources and
their users could potentially be detrimental, as there is a
potential conflict of interest for the providers. They profit
by providing resources to up-and-coming players, but also
wish to maintain dominant positions in their consumer-
facing industries.
3) Environmental Impact: The other major concern is
the ever-increasing carbon footprint from the exponential
growth [4] of the data centres required for Cloud Com-
puting. With the industry expected to exceed the airline
industry by 2020 [6], raising sustainability concerns [5].
The industry is being motivated to address the problem
by legislation [6], [35], the operational limit of power
grids (being unable to power anymore servers in their
data centres) [36], and the potential financial benefits of
increased efficiency [37], [6]. Their primary solution is
the use of virtualisation4to maximise resource utilisation
[39], but the problem remains [40], [41].
While these issues are endemic to Cloud Computing,
they are not flaws in the Cloud conceptualisation, but the
vendor provision and implementation of Clouds [25], [22],
[42]. There are attempts to address some of these concerns,
such as a portability layer between vendor Clouds to avoid
lock-in [43]. However, this will not alleviate issues such as
inter-Cloud latency [44]. An open source implementation
of the Amazon (EC2) Cloud [22], called Eucalyptus [45],
allows a data centre to execute code compatible with
Amazon’s Cloud. Allowing for the creation of private
internal Clouds, avoiding vendor lock-in and providing
information privacy, but only for those with their own data
centre and so is not really Cloud Computing (which by
definition is to avoid owning data centres [1]). Therefore,
vendor Clouds remain synonymous with Cloud Computing
[13], [1], [14], [15]. Our response is an alternative model
for the Cloud conceptualisation, created by combining the
Cloud with paradigms from Grid Computing, principles
from Digital Ecosystems, and sustainability from Green
Computing, while remaining true to the original vision of
the Internet [46].
III. GRID COMPUTING: DISTRIBUTING PROV IS IO N
Grid Computing is a form of distributed computing in
which a virtual super computer is composed from a cluster
4Virtualisation is the creation of a virtual version of a resource, such as a
server, which can then be stored, migrated, duplicated, and instantiated
as needed, improving scalability and work load management [38].
Figure 3. Grid Computing: Typical configuration in which resource
provision is managed by a group of distributed nodes [47]. Green
symbolises resource consumption, and yellow resource provision. The
role of coordinator for resource provision is designated by red, and is
centrally controlled.
of networked, loosely coupled computers, acting in concert
to perform very large tasks [47]. It has been applied
to computationally intensive scientific, mathematical, and
academic problems through volunteer computing, and used
in commercial enterprise for such diverse applications as
drug discovery, economic forecasting, seismic analysis,
and back-office processing to support e-commerce and
web services [47].
What distinguishes Grid Computing from cluster com-
puting is being more loosely coupled, heterogeneous, and
geographically dispersed [47]. Also, grids are often con-
structed with general-purpose grid software libraries and
middleware, dividing and apportioning pieces of a program
to potentially thousands of computers [47]. However, what
distinguishes Cloud Computing from Grid Computing is
being web-centric, despite some of its definitions being
conceptually similar (such as computing resources being
consumed as electricity is from power grids) [9].
IV. DIG ITAL ECOSYSTEMS: DISTRIBUTING CONTROL
Digital Ecosystems are distributed adaptive open socio-
technical systems, with properties of self-organisation,
scalability and sustainability, inspired by natural ecosys-
tems [48], [49]. Emerging as a novel approach to the
catalysis of sustainable regional development driven by
SMEs [50]. Aiming to help local economic actors become
active players in globalisation [51], valorising their local
culture and vocations, and enabling them to interact and
create value networks at the global level [52]. Increas-
ingly this approach, dubbed glocalisation, is being consid-
ered a successful strategy of globalisation that preserves
regional growth and identity [53], [54], [55], and has
been embraced by the mayors and decision-makers of
thousands of municipalities [56]. The community focused
on the deployment of Digital Ecosystems, REgions for
Digital Ecosystems Network (REDEN) [50], is supported
by projects such as the Digital Ecosystems Network of
regions for (4) DissEmination and Knowledge Deployment
(DEN4DEK) [57]. This thematic network that aims to
share experiences and disseminate knowledge to let re-
gions effectively deploy of Digital Ecosystems at all levels
(economic, social, technical and political) to produce real
impacts in the economic activities of European regions
through the improvement of SME business environments.

In a traditional market-based economy, made up of
sellers and buyers, the parties exchange property, while
in a new network-based economy, made up of servers and
clients, the parties share access to services and experiences
[58]. Digital Ecosystems aim to support network-based
economies reliant on next-generation ICT that will extend
the Service-Oriented Architecture (SOA) concept [59]
with the automatic combining of available and applicable
services in a scalable architecture, to meet business user
requests for applications that facilitate business processes.
Digital Ecosystems research is yet to consider scalable re-
source provision, and therefore risks being subsumed into
vendor Clouds at the infrastructure level, while striving for
decentralisation at the service level. So, the realisation of
their vision requires a form of Cloud Computing, but with
their principle of community-based infrastructure where
individual users share ownership [48].
V. GR EE N COMPUTING: GROWING SUSTAI NABLY
Green Computing is the efficient use of computing
resources, with the primary objective being to account for
the triple bottom line5, an expanded spectrum of values
and criteria for measuring organisational (and societal)
success [61]. Given computing systems existed before con-
cern over their environmental impact, it has generally been
implemented retroactively, but is now being considered
at the development phase [61]. It is systemic in nature,
because ever-increasingly sophisticated modern computer
systems rely upon people, networks and hardware. So,
the elements of a green solution may comprise items
such as end user satisfaction, management restructur-
ing, regulatory compliance, disposal of electronic waste,
telecommuting, virtualisation of server resources, energy
use, thin client solutions and return on investment [61].
One of the greatest environmental concerns of the indus-
try is their data centres [41], which have increased in num-
ber over time as business demands have increased, with
facilities housing a rising amount of evermore powerful
equipment [17]. As data centres run into limits related to
power, cooling and space, their ever-increasing operation
has created a noticeable impact on power grids [36]. To the
extent that data centre efficiency has become an important
global issue, leading to the creation of the Green Grid
[62], an international non-profit organisation mandating an
increase in the energy efficiency of data centres. Their
approach, virtualisation, has improved efficiency [40],
[41], but is optimising a flawed model that does not
consider the whole system, where resource provision is
disconnected from resource consumption. For example,
competing vendors must host significant redundancy in
their data centres to manage usage spikes and maintain
the illusion of infinite resources. So, we would argue that
an alternative more systemic approach is required, where
resource consumption and provision are connected, to
minimise the environmental impact and allow sustainable
growth.
5The triple bottom line (people, planet, profit) [60].
VI. COMMUNITY CL OU D
C3 arises from concerns over Cloud Computing, specif-
ically control by vendors and lack of environmental sus-
tainability. The Community Cloud aspires to combine
distributed resource provision from Grid Computing, dis-
tributed control from Digital Ecosystems and sustainability
from Green Computing, with the use cases of Cloud
Computing, while making greater use of self-management
advances from Autonomic Computing. Replacing vendor
Clouds by shaping the underutilised resources of user
machines to form a Community Cloud, with nodes po-
tentially fulfilling all roles, consumer,producer, and most
importantly coordinator, as shown in Figure 4.
Figure 4. Community Cloud: Created from shaping the underutilised
resources of user machines, with nodes potentially fulfilling all roles,
consumer, producer, and most importantly coordinator. Green symbolises
resource consumption, yellow resource provision, and red resource
coordination.
A. Conceptualisation
The conceptualisation of the Community Cloud draws
upon Cloud Computing [20], Grid Computing [9], Digital
Ecosystems [48], Green Computing [63] and Autonomic
Computing [12]. A paradigm for Cloud Computing in the
community, without dependence on Cloud vendors, such
as Google, Amazon, or Microsoft.
1) Openness: Removing dependence on vendors makes
the Community Cloud the open equivalent to vendor
Clouds, and therefore identifies a new dimension in the
open versus proprietary struggle [64] that has emerged in
code, standards and data, but has yet to be expressed in
the realm of hosted services.
2) Community: The Community Cloud is as much a
social structure as a technology paradigm [65], because of
the community ownership of the infrastructure. Carrying
with it a degree of economic scalability, without which
there would be diminished competition and potential sti-
fling of innovation as risked in vendor Clouds.
3) Individual Autonomy: In the Community Cloud,
nodes have their own utility functions in contrast with data
centres, in which dedicated machines execute software as
instructed. So, with nodes expected to act in their own self-
interest, centralised control would be impractical, as with
consumer electronics like game consoles [66]. Attempts
to control user machines counter to their self-interest
results in cracked systems, from black market hardware

modifications and arms races over hacking and securing
the software (routinely lost by the vendors) [66]. In the
Community Cloud, where no concrete vendors exist, it
is even more important to avoid antagonising the users,
instead embracing their self interest and harnessing it for
the benefit of the community with measures such as a
community currency.
4) Identity: In the Community Cloud each user would
inherently possess a unique identity, which combined with
the structure of the Community Cloud should lead to
an inversion of the currently predominant membership
model. So, instead of users registering for each website
(or service) anew, they could simply add the website to
their identity and grant access. Allowing users to have
multiple services connected to their identity, instead of
creating new identities for each service. This relationship
is reminiscent of recent application platforms, such as
Facebook’s f8 and Apple’s App Store, but decentralised
in nature and so free from vendor control. Also, allowing
for the reuse of the connections between users, akin to
Google’s Friend Connect, instead of reestablishing them
for each new application.
5) Graceful Failures: The Community Cloud is not
owned or controlled by any one organisation, and therefore
not dependent on the lifespan or failure of any one
organisation. It therefore ought be robust and resilient to
failure, and immune to the system-wide cascade failures of
vendor Clouds, because of the diversity of its supporting
nodes. When occasionally failing doing so gracefully,
non-destructively, and with minimal downtime, as the
unaffected nodes mobilise to compensate for the failure.
6) Convenience and Control: The Community Cloud,
unlike vendor Clouds, has no inherent conflict between
convenience and control, resulting from its community
ownership providing distributed control, which would
be more democratic. However, whether the Community
Cloud can provide technically quality equivalent or su-
perior to its centralised counterparts is an issue that will
require further research.
7) Community Currency: The Community Cloud
would require its own currency to support the sharing of
resources, a community currency, which in economics is a
medium (currency), not backed by a central authority (e.g.
national government), for exchanging goods and services
within a community [67]. It does not need to be restricted
geographically, despite sometimes being called a local
currency [68]. An example is the Fureai kippu system
in Japan, which issues credits in exchange for assistance
to senior citizens [69]. Family members living far from
their parents can earn credits by offering assistance to
the elderly in their local community, which can then be
transferred to their parents and redeemed by them for local
assistance [69].
8) Quality of Service: Ensuring acceptable quality of
service (QoS) in a heterogeneous system will be a chal-
lenge. Not least because achieving and maintaining the
different aspects of QoS will require reaching critical
mass in the participating nodes and available services.
Thankfully, the community currency could support long-
term promises by resource providers and allow the higher
quality providers, through market forces, to command a
higher price for their service provision. Interestingly, the
Community Cloud could provide a better QoS than vendor
Clouds, utilising time-based and geographical variations
advantageously in the dynamic scaling of resource provi-
sion.
9) Environmental Sustainability: We expect the Com-
munity Cloud to have a smaller carbon footprint than
vendor Clouds, on the assumption that making use of
underutilised user machines requires less energy than the
dedicated data centres required for vendor Clouds. The
server farms within data centres are an intensive form
of computing resource provision, while the Community
Cloud is more organic, growing and shrinking in a symbi-
otic relationship to support the demands of the community,
which in turn supports it.
10) Service Composition: The great promise of service-
oriented computing is that the marginal cost of creating the
n-th application will be virtually zero, as all the software
required already exists to satisfy the requirements of other
applications. Only their composition and orchestration are
required to produce a new application [70], [71]. Within
vendor Clouds it is possible to make services that expose
themselves for composition and compose these services,
allowing the hosting of a complete service-oriented archi-
tecture [20]. However, current service composition tech-
nologies have not gained widespread adoption [72]. Digital
Ecosystems advocate service composability to avoid cen-
tralised control by large service providers, because easy
service composition allows coalitions of SMEs to compete
simply by composing simpler services into more complex
services that only large enterprises would otherwise be
able to deliver [52]. So, we should extend decentralisation
beyond resource provision and up to the service layer, to
enable service composition within the Community Cloud.
B. Architecture
Service Layer
Repository, Composition, Execution
Resource Layer
Computation, Persistence, Bandwidth, Currency
Coordination Layer
Virtual Machine, Identity, Networking, Transactions
Figure 5. Community Cloud Computing: An architecture in which the
most fundamental layer deals with distributing coordination. One layer
above, resource provision and consumption are arranged on top of the
coordination framework. Finally, the service layer is where resources
are combined into end-user accessible services, to then themselves be
composed into higher-level services.
The method of materialising the Community Cloud
is the distribution of its server functionality amongst a
population of nodes provided by user machines, shaping
their underutilised resources into a virtual data centre.
While straightforward in principle, it poses challenges on

many different levels. So, an architecture for C3 can be
divided into three layers, dealing with these challenges
iteratively. The most fundamental layer deals with dis-
tributing coordination, which is taken for granted in ho-
mogeneous data centres where good connectivity, constant
presence and centralised infrastructure can be assumed.
One layer above, resource provision and consumption are
arranged on top of the coordination framework. Easy in the
homogeneous grid of a data centre where all nodes have
the same interests, but more challenging in a distributed
heterogeneous environment. Finally, the service layer is
where resources are combined into end-user accessible
services, to then themselves be composed into higher-level
services.
1) Coordination Layer: To achieve coordination, the
nodes need to be deployed as isolated virtual machines,
forming a fully distributed P2P6network that can provide
support for distributed identity, trust, and transactions.
a) Virtual Machines (VMs): Executing arbitrary code
in the machine of a resource-providing user would require
asandbox7for the guest code, a VM8to protect the
host. The role of the VM is to make system resources
safely available to the Community Cloud, upon which
Cloud processes could be run safely (without danger to
the host machine). Fortunately, feasibility has been proven
with heavyweight VMs such as the Java Virtual Machine,
lightweight JavaScript VMs present in most modern web
browsers, and new approaches such as Google’s Native
Client. Furthermore, the age [76] of multi-core processors9
has resulted in unused and underutilised cores being
commonplace in modern personal computers [78], which
lend themselves well to the deployment and background
execution of Community Cloud facing VMs. Regarding
deployment, users would be required to maintain an active
browser window or tab, or install a dedicated application.
While the first would not require installation privileges,
the later would with the benefit of greater functionality.
However, more likely a hybrid of both would occur, facil-
itating the availability and advantages of each in different
scenarios.
b) Distributed Identity: In distributed systems with
variable node reliability, historical context is logically
required to have certainty of node interactions. Funda-
mental to this context is the ability to identify nodes and
therefore reference previous interactions. However, current
identification schemes have identity providers controlling
6Peer-to-peer (P2P) computing or networking is a distributed applica-
tion architecture that partitions tasks or work loads between service
peers. Peers are equally privileged participants in the application, and
are said to form a peer-to-peer network of nodes [73].
7A sandbox is a security mechanism for safely running programs, often
used to execute untested code, or untrusted programs from unverified
third-parties, suppliers and untrusted users [74].
8A virtual machine is a software implementation of a machine (com-
puter) that executes programs like a real machine [75].
9A multi-core processor is an integrated circuit to which two or more
processors have been attached for enhanced performance, reduced
power consumption, and more efficient simultaneous processing of
multiple tasks [77].
provision. Such as in the DNS10, which while nomi-
nally distributed, remains under centralised control both
technologically and organisationally, permitting numerous
distortions in the network. Including domain squatting11,
abuses by domain registrars [81], subjection to political
control [82], [83] and risks to the infrastructure being
compromised [84]. Identity in the Community Cloud has
to arise naturally from the structure of the network, based
on the relation of nodes to each other, so that it can scale
and expand without centralised control. We can utilise the
property that a large enough identifier-space is unlikely to
suffer collisions. For example, the Git distributed version
control system [85] assigns a universal identifier to each
new commission, without coordination with other repos-
itories. Analogously, assuming each node independently
produces a private-public key pair, the probability of
public key collision is negligible. Also, from the human
identification of nodes we can utilise the property that each
node, despite formal identity, possesses a unique position
in the network, i.e. set of connections to other nodes.
So, combining these two properties provides reasonable
certainty for a distributed identity model where universal
identification can be accomplished without centralised
mediation, but this is still an active area of research.
c) Networking: At this level, nodes should be inter-
connected to form a P2P network. Engineered to provide
high resilience while avoiding single points of control and
failure, which would make decentralised super-peer based
control mechanisms [86] insufficient. Newer P2P designs
[87] offer sufficient guarantees of distribution, immunity
to super-peer failure, and resistance to enforced control.
For example, in the Distributed Virtual Super-Peer (DVSP)
model a collection of peers logically combine to form a
virtual super-peer [87], which dynamically changes over
time to facilitate fluctuating demands.
d) Distributed Transactions: A key element of dis-
tributed coordination is the ability of nodes to jointly
participate in transactions that influence their individual
state. Appropriately annotated business processes can be
executed over a distributed network with a transactional
model maintaining the ACID12 properties on behalf of the
initiator [89]. Newer transaction models maintain these
properties while increasing efficiency and concurrency.
Other directions of research include relaxing these prop-
erties to maximise concurrency [90]. Others still, focus on
distributing the coordination of transactions [87]. A feature
vital for C3, as distributed transaction capabilities are
10 The Domain Name System (DNS) is a hierarchical naming-space for
computers, services, and other resources participating in the Internet. It
translates domain names meaningful to humans into their counterpart
numerical identifiers associated with networking equipment to locate
and address these devices world-wide [79]. So, translating human-
friendly computer hostnames into Internet Protocol (IP) addresses, e.g.
www.example.com translates to 208.77.188.166.
11 Domain squatting (also known as cybersquatting) is registering, traf-
ficking in, or using a domain name in bad faith, with the intent to profit
from the goodwill of a trademark belonging to someone else. The
cybersquatter then offers to sell the domain to the person or company
who owns a trademark contained within the name at an inflated price
[80].
12 ACID (Atomicity, Consistency, Isolation, Durability) is a set of
properties that guarantee transactions are processed reliably [88].

fundamental to permitting multi-party service composition
without centralised mediation.
2) Resource layer: With the networking infrastructure
now in place, we can consider the first consumer-facing
uses for the virtual data centre of the Community Cloud.
Offering the usage experience of Cloud Computing on the
PaaS layer and above, because Cloud Computing is about
using resources from the Cloud. So, Utility Computing
scenarios [91], such as access to raw storage and com-
putation, should be available at the PaaS layer. Access to
these abstract resources for service deployment would then
provide the SaaS layer.
a) Distributed Computation: The field has a success-
ful history of centrally controlled incarnations [92]. How-
ever, C3 should also take inspiration from Grid Computing
and Digital Ecosystems to provide distributed coordination
of the computational capabilities that nodes offer to the
Community Cloud.
b) Distributed Persistence: The Community Cloud
would naturally require storage on its participating nodes,
taking advantage of the ever-increasing surplus on most13
personal computers [94]. However, the method of infor-
mation storage in the Community Cloud is an issue with
multiple aspects. First, information can be file-based or
structured. Second, while constant and instant availability
can be crucial, there are scenarios in which recall times
can be relaxed. Such varying requirements call for a
combination of approaches, including distributed storage
[95], distributed databases [96] and key-value stores [26].
Information privacy in the Community Cloud should be
provided by the encryption of user information when on
remote nodes, only being unencrypted when accessed by
the user, allowing for the secure and distributed storage of
information.
c) Bandwidth Management: The Community Cloud
would probably require more bandwidth at the user
nodes than vendor Clouds, but can take advantage of the
ever-increasing bandwidth and deployment of broadband
[97]. Also, P2P protocols such as BitTorrent [98] make
the distribution of information over networks much less
bandwidth-intensive for content providers, accomplished
by using the downloading peers as repeaters of the infor-
mation they receive. C3 should adopt such approaches to
ensure the efficient use of available network bandwidth,
avoiding fluctuations and sudden rises in demand (e.g. the
Slashdot effect14) burdening parts of the network.
d) Community Currency: An important theme in the
Community Cloud is that of nodes being contributors as
well as consumers, which would require a community cur-
rency (redeemable against resources in the community) to
reward users for offering resources [100]. This would also
allow traditional Cloud vendors to participate by offering
their resources to the Community Cloud to gather consid-
13 The only exception is the recent arrival of Solid-State Drives (SSDs),
popular for mobile devices because of their lack of moving parts,
growing in use as their size and price reach that of traditional Hard
Disk Drives (HDDs) [93].
14 The Slashdot effect, also known as slashdotting, is the phenomenon
of a popular website linking to a smaller site, causing the smaller site
to slow down or even temporarily close due to the increased traffic
[99].
erable community currency, which they can then monetise
against participants running a community currency deficit
(i.e. contributing less then they consume). The relative
cost of resources (storage, computation, bandwidth) should
fluctuate based on market demand, not least because of the
impracticality of predicting or hard-coding such ratios. So,
a node of the network would gather community currency
by performing tasks for the community, which its user
could then use to access resources of the Community
Cloud.
e) Resource Repository: Given that each node pro-
viding resources has a different location in the network
and quality characteristics, a distributed resource reposi-
tory would be required that could respond to queries for
resources according to desired performance profiles. Such
a query would have to consider historical performance,
current availability, projected cost and geographical dis-
tribution of the nodes to be returned. A constraint opti-
misation problem, the results returned would be a set of
nodes that fit the required profile, proportionally to the
availability of suitable nodes.
3) Service Layer: Cloud Computing represents a new
era for service-oriented architectures, making services
explicitly dependent on other resource providers instead
of building on self-sufficient resource locations. C3 makes
this more explicit, breaking down the stand-alone service
paradigm, with any service by default being composed
of resources contributed by multiple participants. So, the
following sections define the core infrastructural services
that the Community Cloud would need to provide.
a) Distributed Service Repository (DSR): The ser-
vice repository of the Community Cloud must provide
persistence, as with traditional service repositories [101],
for the pointers to services and their semantic descriptions.
To support the absence of service-producing nodes during
service execution, there must also be persistence of the
executable code of services. Naturally, the implementation
of a distributed service repository is made easier by the
availability of the distributed storage infrastructure of the
Community Cloud.
b) Service Deployment and Execution: When a ser-
vice is required, but is not currently instantiated on a
suitable node, a copy should be retrieved from the DSR
and instantiated as necessary, allowing for flexible respon-
siveness and resilience to unpredictable traffic spikes. As
nodes are opportunistically interested in executing services
to gather community currency for their users, so developers
should note the resource cost of their services in their
descriptions, allowing for pre-execution resource budget-
ing, and post-execution community currency payments.
Being in a developer’s own interest to mark resource
costs correctly, because over-budgeting would burden their
users and under-budgeting would cause premature service
termination. Additionally, developers could add a subsidy
to promote their services. Remote service execution would
need to be secured against potentially compromised nodes,
perhaps through encrypted processing schemes [102]. Oth-
erwise, such nodes while unable to access a complete
traffic log of the services they execute, could potentially
access the business logic; and we would be replacing
the vendor introspection problem, with an anyone intro-

spection problem. Since delivering a service over large
distances in the network comes at a potentially high cost,
the lack of a central well-connected server calls for a
fundamental paradigm shift, from pull-oriented approaches
to hybrid push/pull-oriented approaches. So, instead of the
pull-oriented approach of supplying services only upon
request [103], service provision should also follow a push-
oriented approach of preemptive deployment to strategi-
cally suitable nodes, including modifying their deployment
profile based on the traffic patterns they face at run time.
c) Programming Paradigm: A key innovation of
Cloud Computing in its PaaS incarnation, is the offer-
ing of a well-specified context (programming paradigm)
within which the services should be executed [20]. The
programming paradigm that produces these services is also
important to C3, because it forms a contract between the
service developers and resource providers. The current
state-of-the-art requires manipulation of source code in
which each line is context dependent, and so a single
intended change may necessitates significant alterations
at different locations in the codebase. A paradigm shift
to declarative generative programming [104] would be
greatly beneficial, avoiding the need to manually manage
cascading changes to the codebase. As the requirements
behind a service would be made explicit and executable,
and being human readable could therefore be manipu-
lated directly as stand-alone artifacts. Additionally, barriers
to service composition would be significantly decreased
[105], beneficial to C3 and beyond.
C. Distributed Innovation
When considering the Community Cloud over time, cur-
rent software distribution models would cause problems.
Should the infrastructure be dependent on a single provider
for updates, they would become a single point of control,
and possibly failure. Entrusting a single provider with the
power to control the evolution of the architecture, even
if they are considered benevolent, risks the development
goals becoming misaligned with the community. There-
fore, the Community Cloud should follow an evolutionary
software distribution model. Extending an already-growing
trend of using distributed code repositories such as Git [85]
and Mercurial [106], over centralised code repositories
such as Subversion [107] and CVS [108]. So, modifica-
tions to services, including infrastructural ones, should be
distributed locally to migrate over the Community Cloud
from where they are deployed, making use of the existing
relationships between users. Users or their nodes (by
default) could even choose to follow the updates that other
trusted peers adopt. Therefore, new versions of a service
would compete with older versions, and where superior
(fitter) would distribute more widely, spreading further
across the Community Cloud. So, updates to services
would permeate through the network, in a distributed but
regulated manner. We could even consider the updates to
services, as the release of patches (modifications), allowing
for frequent, smaller and iterative releases more akin
to an evolutionary software distribution model. Potential
speciation (branching) would encourage developers to co-
ordinate their releases and ensure their patches are viable
across different branches. Obviously, the ability to undo
patches and step back through versions of infrastructural
services would be necessary to maintain the Community
Cloud. Still, without a more granular approach to conflict
resolution from different patching sources, poor developer
relations could risk fragmentation of the codebase and
network. So, an alternative non-centralised software inno-
vation model would be required, such as the declarative
generative programming paradigm [104] mentioned.
VII. INTHE COMMUNITY CL OU D
While we have covered the fundamental motivations
and architecture of the Community Cloud, its practical
application may still be unclear. So, this section discusses
the cases of Wikipedia and YouTube, where the application
of C3 would yield significant benefits, because they have
unstable funding models, require increasing scalability,
and are community oriented.
A. Wikipedia
Wikipedia suffers from an ever-increasing demand for
resources and bandwidth, without a stable supporting
revenue source [109]. Their current funding model re-
quires continuous monetary donations for the maintenance
and expansion of their infrastructure [110]. The alterna-
tive being contentious advertising revenues [109], which
caused a long-standing conflict within their community
[111]. While it would provide a more scalable funding
model, some fear it would compromise the content and/or
the public trust in the content [112]. Alternatively, the
Community Cloud could provide a self-sustaining scalable
resource provision model, without risk of compromising
the content or public trust in the content, because it
would be compatible with their communal nature (unlike
their current data centre model), with their user base
accomplishing the resource provision they require.
Were Wikipedia to adopt C3, it would be distributed
throughout the Community Cloud alongside other services.
With the core operations of Wikipedia, providing web-
pages and executing server-side scripts, being handled as
service requests. Participants would use their community
currency to interact with Wikipedia, performing a search
or retrieving a page, while gaining community currency for
helping to host Wikipedia across the Community Cloud.
More complicated tasks, such as editing a Wikipedia web-
page, would require an update to the distributed storage of
the Community Cloud, achieved by transmitting the new
data through its network of nodes, most likely using an
eventual consistency model [90].
B. YouTube
YouTube requires a significant bandwidth for content
distribution, significant computational resources for video
transcoding, and is yet to settle on a profitable business
model [113], [114]. In the Community Cloud, websites
like YouTube would also have a self-sustaining scalable
resource provision model, which would significantly re-
duce the income required for them to turn a profit.
Were YouTube to adopt C3, it would also be dis-
tributed throughout the Community Cloud alongside other
services. Updates such as commenting on a YouTube
video, would similarly need to propagate through the

distributed persistence layer. So, the community would
provide the bandwidth for content distribution, and the
computational resources for video transcoding, required
for YouTube’s service. The QoS requirements for YouTube
are significantly different to those of Wikipedia, because
while constant throughput is desirable for video stream-
ing, occasional packet loss is tolerable. Also, YouTube’s
streaming of live events has necessitated the services of
bespoke content distribution networks [115], a type of
service for which the Community Cloud would naturally
excel.
We have discussed Wikipedia and YouTube in the Com-
munity Cloud, but other sites such as arXiv and Facebook
would equally benefit. As C3’s organisational model for
resource provision moves the cost of service provision
to the user base, effectively creating a micro-payment
scheme, which would dramatically lower the barrier of
entry for innovative start-ups.
VIII. CONCLUSIONS
We have presented the Community Cloud as an alterna-
tive to Cloud Computing, created from blending its usage
scenarios with paradigms from Grid Computing, principles
from Digital Ecosystems, self-management from Auto-
nomic Computing, and sustainability from Green Com-
puting. So, C3 utilises the spare resources of networked
personal computers to provide the facilities of data centres,
such that the community provides the computing power for
the Cloud they wish to use. A socio-technical conceptual-
isation for sustainable distributed computing.
While the Open Cloud Manifesto [116] is well inten-
tioned, its promotion of open standards for vendor Cloud
interoperability has proved difficult [117]. We believe it
will continue to prove difficult until a viable alternative,
such as C3, is developed. Furthermore, we hope that the
Community Cloud will encourage innovation in vendor
Clouds, forming a relationship analogous to the creative
tension between open source and proprietary software.
In the future we will continue to refine the vari-
ous elements of C3, such as suitable mechanisms for
acommunity currency, distributed alternatives to DNS,
DVSPs, RESTful Clouds, declarative generative program-
ming paradigms, distributed innovation, and the environ-
mental impact of the Community Cloud relative to vendor
Clouds.
ACK NOW LE DG EM EN TS
We would like to thank for comments and helpful
discussions Paulo Siqueira of the Instituto de Pesquisas
em Tecnologia e Inovacao, Eva Tallaksen and Alexander
Deriziotis.
REFERENCES
[1] M. Haynie, “Enterprise cloud services: Deriving business value
from Cloud Computing,” Micro Focus, Tech. Rep., 2009.
[2] R. L. Peter Lucas, Joseph Ballay, “The wrong cloud,” MAYA
Design, Inc, Tech. Rep., 2009. [Online]. Available: http://www.
maya.com/file download/126/The%20Wrong%20Cloud.pdf
[3] M. Armbrust, A. Fox, R. Griffith, A. Joseph, R. Katz,
A. Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica,
and M. Zaharia, “Above the Clouds: A Berkeley view
of Cloud Computing,” University of California, Berkeley,
2009. [Online]. Available: http://d1smfj0g31qzek.cloudfront.net/
abovetheclouds.pdf
[4] J. Hayes, “Cred - or croak?” IET Knowledge Network, Tech.
Rep., 2008. [Online]. Available: http://kn.theiet.org/magazine/
issues/0820/cred-croak- 0820.cfm?SaveToPDF
[5] P. Mckenna, “Can we stop the internet destroying our planet?”
New Scientist, vol. 197, no. 2637, pp. 20–21, 2008.
[6] J. Kaplan, W. Forrest, and N. Kindler, “Revolutionizing data center
energy efficiency,” McKinsey & Company, Tech. Rep., 2008.
[Online]. Available: http://www.mckinsey.com/clientservice/bto/
pointofview/pdf/Revolutionizing Data Center Efficiency.pdf
[7] B. Leiner, V. Cerf, D. Clark, R. Kahn, L. Kleinrock, D. Lynch,
J. Postel, L. Roberts, and S. Wolff, “A brief history of the
internet,” Institute for Information Systems and Computer Media,
Tech. Rep., 2001. [Online]. Available: http://www.iicm.tugraz.
at/thesis/cguetl diss/literatur/Kapitel02/References/Leiner et al.
2000/brief.html?timestamp=1197467969844
[8] J. Scanlon and B. Wieners, “The internet cloud,” The
Industry Standard, Tech. Rep., 1999. [Online]. Available:
http://www.thestandard.com/article/0,1902,5466, 00.html
[9] I. Foster, Y. Zhao, I. Raicu, and S. Lu, “Cloud Computing
and Grid Computing 360-degree compared,” in Grid Computing
Environments Workshop, 2008, pp. 1–10.
[10] T. Foremski, “Sun services CTO says utility computing
acceptance is slow going,” ZDNet, CBS Interactive, Tech. Rep.,
2006. [Online]. Available: http://blogs.zdnet.com/Foremski/?p=33
[11] A. Orlowski, “The Cell chip - how will MS and Intel face the
music?” The Register, Tech. Rep., 2005. [Online]. Available:
http://www.theregister.co.uk/2005/02/03/cell analysis part two/
[12] J. Kephart, D. Chess, I. Center, and N. Hawthorne, “The vision
of autonomic computing,” Computer, vol. 36, no. 1, pp. 41–50,
2003.
[13] G. Gruman and E. Knorr, “What Cloud Computing
really means,” InfoWorld Inc., Tech. Rep., 2008.
[Online]. Available: http://www.infoworld.com/article/08/04/07/
15FE-cloud- computing-reality 1.html
[14] Gartner, “Cloud Computing will be as influential as e-
business,” Gartner, Tech. Rep., 2008. [Online]. Available:
http://www.gartner.com/it/page.jsp?id=707508
[15] P. Gaw, “What’s the difference between Cloud Computing
and SaaS?” Proofpoint, Tech. Rep., 2008. [Online]. Available:
http://blog.fortiva.com/fortivablog/2008/05/what-is- the-dif.html
[16] K. Danielson, “Distinguishing Cloud Computing from
Utility Computing,” ebizQ, Tech. Rep., 2008. [On-
line]. Available: http://www.ebizq.net/blogs/saasweek/2008/03/
distinguishing cloud computing/
[17] M. Arregoces and M. Portolani, Data center fundamentals. Cisco
Press, 2003.
[18] M. Turner, D. Budgen, and P. Brereton, “Turning software into a
service,” Computer, vol. 36, no. 10, pp. 38–44, 2003.
[19] T. Oreilly, “What is Web 2.0: Design patterns and business
models for the next generation of software,” O’Reilly Media,
Tech. Rep., 2008. [Online]. Available: http://www.oreillynet.com/
pub/a/oreilly/tim/news/2005/09/30/what-is-web-20.html
[20] R. Buyya, C. Yeo, and S. Venugopal, “Market-oriented cloud
computing: Vision, hype, and reality for delivering it services
as computing utilities,” in Conference on High Performance
Computing and Communications. IEEE, 2008.
[21] A. Newman, A. Steinberg, and J. Thomas, Enterprise 2.0 Imple-
mentation. McGraw-Hill Osborne Media, 2008.
[22] Amazon, “Amazon Elastic Compute Cloud (EC2),” Amazon
Web Services LLC, Tech. Rep., 2009. [Online]. Available:
http://aws.amazon.com/ec2/
[23] Mosso, “Deploy and scale websites, servers and storage in
minutes,” Rackspace, Tech. Rep., 2009. [Online]. Available:
http://www.mosso.com/
[24] R. Buyya, C. Yeo, and S. Venugopal, “Market-oriented Cloud
Computing: Vision, hype, and reality for delivering it services
as computing utilities,” in High Performance Computing and
Communications. IEEE Press, 2008.
[25] Google, “Google App Engine: Run your web apps on Google’s
infrastructure.” Google, Tech. Rep., 2009. [Online]. Available:
http://code.google.com/appengine/
[26] T. Bain, “Is the relational database doomed?” ReadWriteWeb.com,
2008. [Online]. Available: http://www.readwriteweb.com/archives/
is the relational database doomed.php
[27] F. Chang, J. Dean, S. Ghemawat, W. Hsieh, D. Wallach, M. Bur-
rows, T. Chandra, A. Fikes, and R. Gruber, “Bigtable: A dis-
tributed storage system for structured data,” in USENIX Sympo-
sium on Operating Systems Design and Implementation, 2006.
[28] G. DeCandia, D. Hastorun, M. Jampani, G. Kakulapati, A. Laksh-
man, A. Pilchin, S. Sivasubramanian, P. Vosshall, and W. Vogels,

“Dynamo: Amazon’s highly available key-value store,” in Sympo-
sium on Operating Systems Principles. ACM, 2007, pp. 205–220.
[29] B. Johnson, “Cloud Computing is a trap, warns GNU
founder Richard Stallman,” The Guardian, Tech. Rep., 2008.
[Online]. Available: http://www.guardian.co.uk/technology/2008/
sep/29/cloud.computing.richard.stallman
[30] J. McCabe, Network analysis, architecture, and design. Morgan
Kaufmann, 2007. [Online]. Available: http://books.google.co.uk/
books?id=iddGPgR48 MC
[31] A. Modine, “Web startups crumble under Amazon S3 outage,”
The Register, Tech. Rep., 2008. [Online]. Available: http://www.
theregister.co.uk/2008/02/15/amazon s3 outage feb 2008/
[32] J. Montgomery, “Google Apps sees 99.9relia-
bility”,” Tech.Blorge, Tech. Rep., 2008. [On-
line]. Available: http://tech.blorge.com/Structure:%20/2008/11/
02/google-apps- sees-999- uptime-proves-cloud-reliability/
[33] J. Perez, “Google Apps customers miffed over down-
time,” IDG News Service, Tech. Rep., 2007. [On-
line]. Available: http://www.pcworld.com/businesscenter/article/
130234/google apps customers miffed over downtime.html
[34] Kable, “Carter recommends ’g cloud’ for gov it,” The Register,
2009. [Online]. Available: http://www.channelregister.co.uk/2009/
06/17/government cloud computing/
[35] Environmental Protection Agency, “EPA report to congress on
server and data center energy efficiency,” US Congress, Tech.
Rep., 2007.
[36] R. Miller, “NSA maxes out Baltimore power grid,”
Data Center Knowledge, Tech. Rep., 2006. [Online].
Available: http://www.datacenterknowledge.com/archives/2006/
08/06/nsa-maxes-out-baltimore-power-grid/
[37] K. McIsaac, “The data centre goes green, the CFO saves money,”
Intelligent Business Research Services, Tech. Rep., 2007.
[38] C. Wolf and E. Halter, Virtualization: from the desktop to the
enterprise. Apress, 2005.
[39] R. Talaber, T. Brey, and L. Lamers, “Using virtualization to
improve data center efficiency,” The Green Grid, Tech. Rep., 2009.
[40] K. Brill, “The invisible crisis in the data center: The economic
meltdown of Moore’s law,” Uptime Institute, Tech. Rep., 2007.
[41] J. Brodkin, “Gartner in ‘green’ data centre warning,” Techworld,
2008. [Online]. Available: http://www.techworld.com/green-it/
news/index.cfm?newsid=106292
[42] Microsoft, “Azure services platform,” Micrsoft, Tech. Rep., 2009.
[Online]. Available: http://www.microsoft.com/azure/
[43] C. Metz, “The Meta Cloud - flying data centers enter fourth
dimension,” The Register, Tech. Rep., 2009. [Online]. Available:
http://www.theregister.co.uk/2009/02/24/the meta cloud/
[44] T. Kulmala, “The cloud’s hidden lock-in: Latency,” Archivd,
2009. [Online]. Available: http://blog.archivd.com/1/post/2009/
04/the-clouds- hidden-lock- in-latency.html
[45] D. Nurmi, R. Wolski, C. Grzegorczyk, G. Obertelli, S. Soman,
L. Youseff, and D. Zagorodnov, “The Eucalyptus open-source
cloud-computing system,” in Cloud Computing and Its Applica-
tions, 2008.
[46] J. Abbate, Inventing the internet. MIT press, 1999.
[47] I. Foster and C. Kesselman, The grid: blueprint for a new
computing infrastructure. Morgan Kaufmann, 2004.
[48] G. Briscoe and P. De Wilde, “Digital Ecosystems: Evolving
service-oriented architectures,” in Conference on Bio Inspired
Models of Network, Information and Computing Systems. IEEE
Press, 2006. [Online]. Available: http://arxiv.org/abs/0712.4102
[49] G. Briscoe, “Digital ecosystems,” Ph.D. dissertation, Imperial
College London, 2009.
[50] L. Rivera Le´
on. Regions for Digital Ecosystems Network
(REDEN). [Online]. Available: http://reden.opaals.org/
[51] P. Dini, G. Lombardo, R. Mansell, A. Razavi, S. Moschoyiannis,
P. Krause, A. Nicolai, and L. Rivera Le ´
on, “Beyond interop-
erability to digital ecosystems: regional innovation and socio-
economic development led by SMEs,” International Journal of
Technological Learning, Innovation and Development, vol. 1, pp.
410–426, 2008.
[52] F. Nachira, A. Nicolai, P. Dini, M. Le Louarn, and L. Rivera Le´
on,
Eds., Digital Business Ecosystems. European Commission, 2007.
[53] R. Robertson, “Globalisation or glocalisation,” Journal of Inter-
national Communication, vol. 1, pp. 33–52, 1994.
[54] E. Swyngedouw, “The mammon quest. ‘Glocalisation’, interspa-
tial competition and the monetary order: the construction of new
scales,” in Cities and regions in the new Europe: The Global-
local Interplay and Spatial Development Strategies, M. Dunford
and G. Kafkalas, Eds. Wiley, 1992, pp. 39–67.
[55] H. Khondker, “Glocalization as globalization: Evolution of a
sociological concept,” Bangladesh e-Journal of Sociology, vol. 1,
pp. 1–9, 2004.
[56] Glocal Forum and CERFE, “The glocalization manifesto,”
The Glocal Forum, Tech. Rep., 2004. [Online]. Avail-
able: http://www.glocalforum.org/mediagallery/mediaDownload.
php?mm=/warehouse/documents/the glocalization manifesto.pdf
[57] L. Rivera Le´
on. Digital Ecosystems Network of Regions
for Dissemination and Knowledge Deployment (DEN4DEK).
[Online]. Available: http://www.den4dek.org
[58] P. Delcloque and A. Bramoull ´
e, “DISSEMINATE, an initial im-
plementation proposal: a new point of departue in call for the
‘year 01’?” ReCALL, vol. 13, pp. 277–292, 2001.
[59] E. Newcomer and G. Lomow, Understanding SOA with web
services. Addison-Wesley, 2005.
[60] J. Elkington, Cannibals with forks: the triple bottom line of 21st
century business. New Society Publishers, 1998.
[61] J. Williams and L. Curtis, “Green: The new computing coat of
arms?” IT PROFESSIONAL, pp. 12–16, 2008.
[62] T. G. Grid, “About the green grid,” The Green Grid, Tech.
Rep., 2009. [Online]. Available: http://www.thegreengrid.org/
about-the- green-grid
[63] J. Harris, Green Computing and Green It Best Practices on
Regulations and Industry. Lulu.com, 2008.
[64] J. West and J. Dedrick, “Proprietary vs. open standards in the
network era: An examination of the linux phenomenon,” in System
Sciences. IEEE Press, 2001, p. 10.
[65] Y. Benkler, “Sharing nicely: on shareable goods and the emer-
gence of sharing as a modality of economic production,” The Yale
Law Journal, vol. 114, no. 2, pp. 273–359, 2004.
[66] J. Grand, F. Thornton, A. Yarusso, and R. Baer, Game Console
Hacking: Have Fun While Voiding You Warranty. Syngress Press,
2004.
[67] T. Greco, Money: Understanding and creating alternatives to legal
tender. Chelsea Green, 2001.
[68] A. Doteuchi, “Community currency and NPOs- A model for
solving social issues in the 21st century,” Social Development
Research Group, NLI Research, 2002. [Online]. Available: http://
www.nli-research.co.jp/english/socioeconomics/2002/li0204a.pdf
[69] B. Lietaer, “Complementary currencies in japan today: History,
originality and relevance,” International Journal of Community
Currency Research, vol. 8, pp. 1–23, 2004.
[70] Q. Tang, “Economics of web service provisioning: Optimal market
structure and intermediary strategies,” Ph.D. dissertation, Univer-
sity of Florida, 2004.
[71] G. Modi, “Service oriented architecture & web 2.0,” Guru Tegh
Bahadur Institute of Technology, Tech. Rep., 2007. [Online].
Available: http://www.gsmodi.com/files/SOA Web2 Report.pdf
[72] B. Violino. (2007) How to navigate a sea of SOA standards.
[Online]. Available: http://www.cio.com/article/104007/How to
Navigate a Sea of SOA Standards
[73] R. Schollmeier, “A definition of peer-to-peer networking for the
classification of peer-to-peer architectures and applications,” in
International Conference on Peer-to-Peer Computing. IEEE
Press, 2002, pp. 101–102.
[74] M. Bishop, Computer Security. Addison-Wesley, 2004.
[75] I. Craig, Virtual machines. Springer, 2006.
[76] D. Geer, “Chip makers turn to multicore processors,” IEEE
Computer, vol. 38, no. 5, pp. 11–13, 2005.
[77] M. Zelkowitz, Advances in Computers: Architectural Issues. Aca-
demic Press, 2007.
[78] B. Posey, “Multi-core processors: Their implication
for Windows,” TechTarget, Tech. Rep., 2007. [On-
line]. Available: http://searchwindowsserver.techtarget.com/tip/0,
289483,sid68 gci1248527, 00.html
[79] P. Mockapetris and K. Dunlap, “Development of the domain name
system,” Computer Communication Review, vol. 18, no. 4, pp.
123–133, 1988.
[80] M. Maury and D. Kleiner, “E-commerce, ethical commerce?”
Journal of Business Ethics, vol. 36, no. 1, pp. 21–31, 2002.
[81] G. Lyon, “Exposing the many reasons not to trust godaddy with
your domain names,” NoDaddy.com, Tech. Rep., 2009. [Online].
Available: http://nodaddy.com/
[82] M. Mueller, Ruling the root: Internet governance and the taming
of cyberspace. MIT press, 2002.
[83] (2006) Chinese walls. [Online]. Available: http://www.economist.
com/business/displaystory.cfm?story id=5582257
[84] D. Goodin, “DNS patch averts doomsday scenario,” The Register,
Tech. Rep., 2008. [Online]. Available: http://www.theregister.co.
uk/2008/08/06/kaminsky black hat/

[85] S. Chacon, “About Git,” GitHub, Tech. Rep., 2009. [Online].
Available: http://git-scm.com/about
[86] J. Risson and T. Moors, “Survey of research towards robust peer-
to-peer networks: Search methods,” Computer Networks, vol. 50,
pp. 3485–3521, 2006.
[87] A. Razavi, S. Moschoyiannis, and P. Krause, “A scale-free busi-
ness network for digital ecosystems,” in IEEE Conf. on Digital
Ecosystems and Technologies, 2008.
[88] T. Haerder and A. Reuter, “Principles of transaction-oriented
database recovery,” ACM Computing Surveys, vol. 15, no. 4, pp.
287–317, 1983.
[89] A. Fox, S. Gribble, Y. Chawathe, E. Brewer, and P. Gauthier,
“Cluster-based scalable network services,” ACM SIGOPS Operat-
ing Systems Review, vol. 31, no. 5, pp. 78–91, 1997.
[90] W. Vogels, “Eventually consistent,” ACM Queue, vol. 6, 2008.
[91] M. Rappa, “The utility business model and the future of computing
services,” IBM Systems Journal, vol. 43, no. 1, pp. 32–42, 2004.
[92] H. Attiya and J. Welch, Distributed computing: fundamentals,
simulations, and advanced topics. Wiley-Interscience, 2004.
[93] C. Mellor, “Ssd and hdd capacity goes on embiggening,” The
Register, Tech. Rep., 2009. [Online]. Available: http://www.
theregister.co.uk/2009/01/09/ssd and hdd capacity increases/
[94] M. Daley, “Software bloat,” MattsComputerTrends.com, Tech.
Rep., 2009. [Online]. Available: http://www.mattscomputertrends.
com/softwarebloat.html
[95] P. Yianilos and S. Sobti, “The evolving field of distributed
storage,” IEEE Internet Computing, vol. 5, pp. 35–39, 2001.
[96] H. Garcia-Molina, J. Ullman, and J. Widom, Database Systems:
The complete book. Prentice Hall, 2008.
[97] “Broadband growth and policies in OECD countries,” Organisa-
tion for Economic Co-operation and Development, Tech. Rep.,
2008.
[98] B. Cohen, “Incentives build robustness in BitTorrent,” in Workshop
on Economics of Peer-to-Peer Systems, vol. 6, 2003.
[99] S. Adler, “The Slashdot effect: An analysis of three internet
publications,” Linux Gazette, vol. 38, 1999.
[100] D. Turner and K. Ross, “A lightweight currency paradigm for the
p2p resource market,” in International Conference on Electronic
Commerce Research, 2004.
[101] M. Papazoglou, “Service-oriented computing: concepts, charac-
teristics and directions,” in International Conference on Web
Information Systems Engineering, T. Catarci, M. Mecella, J. My-
lopoulos, and M. Orlowska, Eds. IEEE Press, 2003, pp. 3–12.
[102] C. Gentry, “Fully homomorphic encryption using ideal lattices,” in
Symposium on Theory of computing. ACM, 2009, pp. 169–178.
[103] M. Singh and M. Huhns, Service-Oriented Computing: Semantics,
Processes, Agents. Wiley, 2005.
[104] A. Marinos and P. Krause, “What, not how: A generative approach
to service composition,” in Digital Ecosystems and Technologies
Conference. IEEE Press, 2009.
[105] ——, “Using sbvr, rest and relational databases to develop in-
formation systems native to the digital ecosystem,” in Digital
Ecosystems and Technologies Conference. IEEE Press, 2009.
[106] S. Arseanrapoj, “Mercurial: Source control management system,”
Mercurial, Tech. Rep., 2009. [Online]. Available: http://mercurial.
selenic.com/wiki/
[107] Tigris, “Subversion: open source version control sys-
tem,” Tigris.org, Tech. Rep., 2009. [Online]. Available:
http://subversion.tigris.org/
[108] D. Price, “Cvs - concurrent versions system,” nongnu.org, Tech.
Rep., 2009. [Online]. Available: http://www.nongnu.org/cvs/
[109] Heebie Blog, “Wikipedia Fundraising: The real truth,” Heebie
Intuitive Design, Tech. Rep., 2009. [Online]. Available: http:
//blog.heebie.co.uk/wikipedia-fundraising- real-truth
[110] A. Modine, “Wales’ personal begging earns last $2m,” The
Register, Tech. Rep., 2009. [Online]. Available: http://www.
theregister.co.uk/2009/01/02/wikipedia fundraising 2m jan 2/
[111] W. Roelf, “Wikipedia founder mulls revenue options,” Reuters,
Tech. Rep., 2007. [Online]. Available: http://www.reuters.com/
article/internetNews/idUSL1964587420070420
[112] H. Leslie, “Wikipedia to run out of money?” Digital-Lifestyles,
Tech. Rep., 2007. [Online]. Available: http://digital-lifestyles.
info/2007/02/12/wikipedia-to- run-out- of-money/
[113] C. Metz, “Google accused of avoiding youtube revenues,” The
Register, 2009. [Online]. Available: http://www.theregister.co.uk/
2009/06/18/google youtube loses/
[114] D. Silversmith, “Google losing up to $1.65m a
day on youtube,” Internet Evolution, 2009. [Online].
Available: http://www.internetevolution.com/author.asp?section
id=715&doc id=175123&
[115] M. Arrington, “Google relies on akamai to stream youtube
live; 700,000 concurrent viewers,” TechCrunch, 2008.
[Online]. Available: http://www.techcrunch.com/2008/11/22/
google-relies- on-akamai- to-stream- youtube-live-700000- concurrent-viewers/
[116] “Open cloud manifesto,” OpenCloudManifesto.org, Tech. Rep.,
2009. [Online]. Available: http://www.opencloudmanifesto.org/
Open%20Cloud%20Manifesto.pdf
[117] C. Metz, “What’s an open cloud? the manifesto’s not telling,”
The Register, 2009. [Online]. Available: http://www.theregister.
co.uk/2009/03/31/amazon on cloud manifesto/
[118] C. Hewitt, “ORGs for scalable, robust, privacy-friendly client
Cloud Computing,” IEEE Internet Computing, vol. 12, no. 5, 2008.
[119] A. Avram, “Architecting for green computing,” InfoQ.com,
2008. [Online]. Available: http://www.infoq.com/news/2008/12/
Architecture-Green- Computing
[120] A. Weiss, “Computing in the clouds,” netWorker, vol. 11, pp. 16–
25, ACM Press, 2007.
[121] G. Briscoe and A. Marinos, “Digital ecosystems in the clouds:
Towards community cloud computing,” in Digital Ecosystems
and Technologies Conference. IEEE Press, 2009. [Online].
Available: http://arxiv.org/abs/0903.0694