Designing of Cloud-based Virtual Factory Information System

pullfarmInternet και Εφαρμογές Web

3 Νοε 2013 (πριν από 3 χρόνια και 5 μήνες)

50 εμφανίσεις

Designing of Cloud
-
based Virtual Factory Information System




Designing of Cloud
-
based Virtual Factory Information System


Yuqiuge Hao

1
*
,
Rafael Karbowski
2
,
Ahm Shamsuzzoha
1
, Petri Helo
1

1

Department of Production


2
Ascora GmbH

University
of
Vaasa



D
-
27777 Ganderkesee, Germany

65
101
,

Vaasa
,

Finland

A
BSTRACT

In the manufacturing industry, customers’ requirements vary all the time, a way that to increase capacity and add
capabilities of factories without investing in new infrastructure becomes

essential.
An advanced information
management system to share valuable
information and
knowledge among collaborative factories

is
demanded
.

The
concept of “Cloud” can encompasses subscription
-
based or pay
-
per
-
use service that, in real time over the Internet,
extends factories existing capabilities.
Cloud Storage can be used to share data is a flexible manner.
With such
perspective, a cloud
-
bas
ed Virtual Factory Information System (CloudVFIS) desig
n is proposed in this research.

This new system will provide a concrete tool for SMEs (Small and Medium Enterprises) to realize the integration of
factories based on the idea of Virtual Factor
y
. In thi
s paper, the CloudVFIS architectural framework and the cloud
storage

in manufacturing management are illustrated.

1.

I
NTRODUCTION


Under the pressures of g
lobal competition
,
there is a need for highly flexible business relationship between
enterprises.
M
an
y

enterprises

concentrate on their core competencies while
adopt themselves to
participate in
emerging interenterprise formations following the virtual enterprise (VE) paradigm

[1]
. VE is a form that offers high
flexibility, agility and resilience for ente
rprises to survive and prosper in the globalized economy.

Although the concept of VE has already existed over decades, there is a lack of efficient IT systems to support its
inherent functionalities.

Many researchers proposed
different approaches to estab
lish
VE
,

but
the
challenges with VE
are flexibility, adaptability, and agility
[2]
.
Especially for Sm
all and Medium sized

manufacturing companies, they are
lack of
efficient communication

when cooperate with other
enterprises
. T
herefore, it is very
necessar
y to find a
solution to support and improve
the
VE
activities.


Cloud computing is a new overloaded IT term.
With the advent of cloud computing, more and more business are
taking advantages of cloud computing.
The flexibility of
c
loud
c
omputing mak
es it easier for companies to scale their
services

according to the user demands.
One major characteristics of cloud computing is l
oad balancing
, which

means
the computation loads get balanced on
-
the
-
fly as the number of requests increase or decrease
[3]
.
This load balancing
service maximizes throughput, minimizes response time, and avoid overload
, therefore it ensures the system
performance
[3]
.

In the business environment, our concern is: h
ow could the concepts of Virtual Enterprise and Cloud Computing be

deployed and
integrated
?
A

new strategy for
Cloud based
Virtual Factory Information System (VFIS) design is
proposed in this resear
ch. This

Cloud
VFIS design aims at setting up process
-
based collaboration network, and
leveraging the information exchange be
tween different manufactories

by utiliz
ing different
Cloud Storage
s
.

The
Storage provided by cloud model is named as Storage as a Service [4]. There are several different types of Cloud
Storage Services. The major differences among these different Cloud St
orages are how the customer uses the storages
and how the users access the storages [3].
The main focus in this paper is to introduce a new
methodology

to use Cloud
Storage to support the Cloud based Virtual Factory Information System

and to support the vi
rtual factory
activities
.

In this paper
,
CloudVFIS
reduces

the limits of collaboration between SMEs in the manufacturing domain. This is
performed by providing companies with tools and methods to interact with other companies within virtual factory and
distributed manufacturing processes.

The rest of this paper
is organized as follows: S
ection

2

describes some related work

and

existing literature
;
Section 3 shows how

to
design

CloudVFIS
;
the following
section illustrates

how to implement Cloud storage by an
example. T
he concluding section summarized the entire
pa
per
.




*

Corresponding author: Tel.: (
358
)

2904498306
; Fax: (
358
)
6 324 8467
; E
-
mail:
haoyuqiu
@
uwasa.fi


Flexible Automation and Intelligent Manufacturing, FAIM2013


2.

L
ITERATURE
R
EVIEW

2.
1

V
IRTUAL
F
ACTORY

AND
I
NFORMATION
S
YSTEM

The term of V
irtual
E
nterprise

(VE)

is

used to form a

temporary collaborative network of independent enterprises
.
Many geographically dispersed enterprises often
structure

this
type

of business collaboration to mee
t various market
objects agile
.
The goal of VE is

to
achieve a

particular business
requirement

and obtain more
opportunity
[1]
,
[5]

for
small and medium enterprises
.
The
benefit

of
enterprises
cooperat
ion in the VE is to en
sure
the enterprises

can

focus
on their own core business

while

leverage

their
competencies, resources or the skills.

Industrial manufacturing
is a typical application area of VE concept
, which also our focus in this paper.
In this
turbulent global econom
ic environment
,

the manufacturing processes can be hardly accomplished by only one
manufacturer
, but
multiple manufacturers
.

Manufacturers must be viewed in the context of their contribution to the
total value chain

[5]
.

It’s very important for
manufacture
rs

to

focus on their core competences

and
join efforts with
others
,

in order to fulfill the new products’ requirements
[6]
.

Therefore, we
label

this particular virtual enterprise as
Virtual Factory (VF).
More
specifically
,
VF is dynamic, ad
-
hoc and
temporary
,

and exists only for the lifetime of a
specific business opportunity in the manufacturing.

Each manufacturer in VF is independent operation agency, but the
management of VF is logic concentrated and physical distributed
[7]
.


The
success

of VF i
s highly relying on Information System. Advances in modern technologies, such as internet,
workflow management system etc., have made
Information System
is
possible to enable enterprises to cooperate with
each other
[8]
.

VE structure

is

information system
centered. The core competence which has to be developed is an
infor
mation system working with different organizations.

However, there

are many challenges related to technology
when implement Information System to support
VE
.
The structure of VE has to face

different technical constraints
[9]
.
Particular
ly for VF the challenges could be:

1) selection of proper manufacturing enterprises; 2) rapid integration of
the manufacturing processes; 3)
d
ynamic
reconfiguration

of system; 4)
c
hange introduced by various
factors such as
joining and dropping of partners, market and context changes.


2.
2

D
ATA
S
TORAGE

OF

V
IRTUAL
F
ACTORIES

I
NFORMATION
S
YSTEM

On
e

issue in realizing
VE

is the information/data exchange among different application systems in th
is

virtual
activities
environment
[10]
.

ICT support systems have to allow enterprises to share information, by guaranteeing
data
-
consistency and establishing synchronized and collaborative processes
[11]
.

The difficulty and problem is w
hen

different actors a
re working together at

the

same time, how to handle real
-
time communication [
9]
.
Information/data
exchange between different software tools is a common problem.
Efficient

data management is the most critical issue
to obtain the agility
and it
improv
es

the
competitiveness of virtual enterprises

[10]
.

Cloud computing gains a lot of attention currently.
M
ore and more business are taking advantages of cloud
computing.

In this

virtual environment
,

users
can
access computing
resources remotely

and all the required data is
transferred throughout the cloud
[12]
.
According to the definition of the National Institute of Standards and
Technology (NIST), the cloud
-
based services can be divided into three layers: Software
-
as
-
a
-
Service (SaaS),
Platform
-
as
-
a
-
Service (PaaS), and Infrastructure
-
as
-
a
-
Service (IaaS)

[13]
.

C
loud storage
belongs to IaaS layer and it
is
becoming a very popular research field. One reason is that it’s the basic level of the whole cloud system to support the
funct
ions of other lay
ers above it

(SaaS, PaaS)
. Another reason is more and more data
-
intensive applications are being
attracted by the clouds.

Cloud storage can provide high scalability, availability, fault tolerance, security, and cost
-
effective data services
for those applic
ations

[14]
.

Typically

in

the

cloud storage,
users will know neither the exact location of their data nor
the other sources of the data collectively stored with theirs.

C
loud
-
based

data storages

provide several advantages in
contrast to former data centre
solutions.

Therefore, cloud
-
based storage solutions are applied within
this paper

for the
catalog
ing and processing

various

data, like the incorporation of
eternal system’s data, a
nd product
ion

status
information
,

inter
-
organizational

data
, etc.

This way,
an overall repository for manufacturing process
-
related data can
be provided.

3.

C
LOUD STORAGE IN
C
LOUD
VFIS

In this paper, we propose a CloudVFIS (
Cloud
-
based
Virtual Factory Information System).
In the virtual factory
information system, t
echnology used to communicate data must be cost
-
effective, flexible, and portable
[15]
.
Since

the
considerable
benefits of cloud
data
storage
,
it

plays

an important role in
this Cloud
VFIS

implementation
.


Designing of Cloud
-
based Virtual Factory Information System




3.1

C
LOUD BASED
VFIS

WORKING ENVIRONMENT

The work
ing environment for a virtual factory proposed in Figure 1.

This solution
help
s

virtual factory and the
cooperating
manufacturing enterprises move beyond existing operational limitations by providing concrete tools and
approaches for leveraging the informa
tion exchange between
enterprises
.
The centralized control enterprise is named
as manufacturing
broker. T
hey are responsible to
collect customer orders, design process model, and describe the
manufacturing process and then assign
appropriate

manufacturing
enterprises (in this scenario, they are called as
partner factories) to accomplish the manufacturing processes.

Therefore t
he
clients in this Cloud
VFIS are two parts:
brokers who propose the virtual factory

and

partner factories
.
Moreover, all the information storing and data
performing are taking place in cloud
storage
.

Figure 1: Cloud VFIS working environment

In order to satisfy orders in a short time and avoid bottlenecks
,

the

controlling
over the

whole

manufacturing

process ca
rried out
in

a

form of

virtual factory.

The brokers benefit

from this
Cloud
VFIS by
following

activities
:



Brokers can describe the (semantic) description of companies, their production facilities, and products
.



Brokers can design a process to produce specif
ic products and describe the constraints, e.g. regarding
environmental and ethical questions, lead time, costs.



Brokers can assign different partner factories to each process step. This
Cloud
VFIS can suggest the best fit
partner factory to each step in ord
er to optimize the manufacturing process.



Brokers can simulate the well designed process and then execute the process.



Information about the current status of the manufacturing process is given in real
-
time. Furthermore,
information from different ICT sys
tems of the partner factories is also integrated into process monitoring.



When unforeseen events happen and change the outcome of the manufacturing process (e.g. delay and costs
increase), an automatic adaption of the process will take place.

Based on
this scenario described, a new architecture to implement all these functionalities is designed

in the
following section
.

3.2

C
LOUD BASED
VFIS

A
RCHITECTURE

Three
-
tier architecture model is the fundamental framework for virtual enterprise information system.

In both
research [11] and [7], authors propose three
-
tier model architecture to implement the Virtual Enterprise Information
System.

However,
the traditional three
-
tier architecture is suitable mostly for applications with a predictable number
of users, f
ollowing a small number of usage patterns and a reduced number of load spikes
[16]
.

In other words, this
architecture runs into problems with the need for high scalability and elasticity of modern web applications such as for
virtual factory implementation
. Furthermore, traditional web applications use relational databases for their data tier.
This database system is difficult to scale or to replace in case of failure and any change in the database schema requires
some downtime. Also, performing queries on
these databases is slow
[16]
.
In order to solve these issues in three
-
tier
architecture, we provide a new solution. I
n this paper, this three
-
tier architecture needs to be modified with two
additional services components. The architecture is presented as F
igure 2.


Flexible Automation and Intelligent Manufacturing, FAIM2013















Figure 2: Overview of Cloud VFIS Architecture



Presentation Layer: this layer remains the same functionalities as traditional presentation l
ayer to interact
with users. This user interface can be customized based on users different ne
eds and users can use any devices
with web browser to access the Cloud based VFIS.

The user interface for brokers and partner factories are
different.



Business Layer: this layer

consists of business rules. It is responsible for implementing the manufactur
ing
process
. It contains six distributed components:
Process Design
,
Process Simulation
,
Process Optimization
,
Process Execution
,
Process Monitoring

and
Process Adaptation
.

Although these components are designed
separately, they interact with each other.

T
h
e interplay of these

components
is
used to

manage the
manufacturing process in order to fulfill the customer orders. These components mainly used by brokers, but
when signing the partner factories to the manufacturing process, the system will send notification to the
partner factories. Although th
e partner factories don’t have
active interact with these components, they are
important part of
the manufacturing process
.



Data Layer:
The data layer contains cloud storage. This Cloud Storage serves as central data storage.
It allows
the business layer
to
pass by
different types of data (binary, semi
-
structured,
structured and
semantic) and
will ensure high scalability for data storage processes. Within this
cloud storage
, scalability and reliability
both have a higher priority than speed.

Because there
are many different types of data in
Cloud
VFIS, the
cloud storage should support all of them b
y

different cloud providers.



Load Balancer Service: This service distributes user requests to different components in the business layer
instances in order to avoi
d overload and minimize response time.



Message Routing Service: This service is needed which connects the different tiers and which manages the
communications between
each tier
. This service

routes valuable information and ensure the scalability
.

4.

C
LOUD
S
TORAGE
I
MPLEMENTATION

CloudVFIS
applies the cloud storage to support virtual factory information system. It allows manufacturers to
share information about production/products over the cloud platform.
I
n this
section
, we only focus on cloud storage
implem
entation.

The
data layer

support
s

several types of data
. For instance,
semi
-
structured

data storage

such as
XML or JSON
data
,

structured

data
used internally by
different components in business layer
,
semantic

data for semantic company
descriptions and also
binary

data
, which is

documents such as specifications.

Therefore, different cloud storages are
needed to different types of data.

This Cloud Storage will be based on the concept of buckets, which are
specific
isolated storage spaces managing data for different data type.

These buckets can be thought as independent databases
Designing of Cloud
-
based Virtual Factory Information System




to store and retrieve different types of data
in
different databases. Thereby, the data types decide their databases and
how it st
ores the data.

Within this solution
, a set of
four

different bucket types for semi
-
structured,
structured
,

semantic and binary data
will be implemented
. Designing in this way can ensure the flexibility. If other

bucket types

are needed
,
such as
a SQL
buck
et,
they
can be added easily
.
The Cloud Storage will support a basic set of CRUD (create, read, update, delete)
operations for all bucket types in a suitable data format (e.g. OData for structured data).

4.1

C
LOUD
S
TORAGE
S
TRUCTURE

In order to realize this

integrated
Cloud Storage, it is needed to identify adequate technologies to store structured,
semi
-
structured
,

binary
and semantic data.
One unique feature of the business model in cloud is that different services
are provided by multiple operators
[4]
.
T
herefore, different cloud storage providers will be selected to provide
different types of cloud storages.
For

managing

those data types, a lot of proven
technologies

exist. In this paper, the
well tested and good performance technologies are selected. It is
presented

in Table 1.

Table 1: Description of different bucket types and the technology selection

Bucket Types


Type of Data


Requirements


Selected
Technology


Structured

Typical application data
such as setting or
administration data

Table based structure on top of relational databases

MySQL

Semi
-
structured

Typical data in a document
-
oriented way
without a fixed data schema
, such as XML
or JSON data


It executes the semi
-
structured (for example for NoSQL
databases) queries and returns the results to the Cloud Storage
Facade.

MangoDB


Binary

A document
-
centric storage for binary data
(e.g. Store videos, PDFs, etc.)

Queries will be based on the document name or ID, e.g. by
requesting the content of the document

company
description.pdf”

Amazon S3

Semantic

Storage of semantic information, e.g. for
managing semantic factory descriptions

Queries will be based on a semantic query language such as
SparQL

Sesame


Based on the selection of technology on different types of bucket, the Cloud Storage can be decomp
osed

into a
detailed structure. Figure 3 illustrates the structure of Cloud Storag
e.



Message Routing: It is used to realize the communication with
in Cloud
VFIS
.

It receives message from other
components in business layer to this data layer.



Cloud Storage Facade:
this is the message interface
and virtual controller
of
this
Cloud Storage.
It manages
the buckets, interprets the messages and executes the commands sent in the message. Additionally, it checks,
whether the data has to be transformed and if the needed access rights are granted. To achieve this it uses the
Query Tra
nslator and the Access Control.




Query Translator: It is used to convert the data from the messaging format
into

the specific database
query format and back.



Access Control: it is used to check if components in business layer are authenticated to access a
specific
bucket. It also checks whether the users have the rights to access specific binary data. Access Control
List will be stored in the Semi
-
Structured Data Storage.



Buckets: Buckets are independent data storages that can be created and used by compon
ents. Depending on
the type they are realized as own database instances, separate tables or keys with are specific prefix. Each
component can create an own separate bucket, so the Cloud Storage has to manage several buckets of each
type, as it can be seen
as an example in

Figure 3, where two binary buckets exist

in Binary Data Storage
.







Flexible Automation and Intelligent Manufacturing, FAIM2013














Figure 3: Cloud Storage Structure

4.2

C
REATE
B
UCKET

This is an example when a component
, for instance Process Design,

in business layer
sends a message

to data layer

to create a binary bucket.
The Process Design component sends a message with the bucket type, component identifier
and the command to create a bucket to the Cloud Storage. Then the
responsible

Cloud Storage provider creates the
bucket and se
ts the corresponding value in the ACL that the component can access the bucket. In the end, the Cloud
Storage sends the message with the bucket ID back to the component that sends the message to create the bucket
(Figure 4).












Figure 4: Use case

for Bucket Creation

The Message Routing Service throws an event that a message has been received. The Cloud Storage Facade checks
this message, which kind of bucket has to be created. A Binary bucket will be created in Amazon S3 and returns a
BucketID. Th
en the Cloud Storage Facade sets the needed write and read rights for the component that has sent the
message and it returns the BucketID to it. For instance, Process Optimization component has the right to access
Process Design’s buckets with read and wri
te access. But some other components only have read
-
only access. Also,
Process Design and Process Execution can share a bucket to store the process models (Figure 5).




Designing of Cloud
-
based Virtual Factory Information System















Figure

5
:

Sequence diagram for bucket creation

This method will create a

new bucket for an application for
managing data:

public String CreateBucket( String bucketType, String componentID,


Boolean publicRead, Boolean publicWrite )

Parameters

bucketType: Type of the bucket that should be created
.


values
=“Structured”,


SemiStructured”,


Binary”,


Semantic”

componentID: Identifier of the component that wants to create the bucket


It will be provided by the Message Routing component.

publicRead: Flag, for public read access. Default: false

publicWrite:
Flag, for public write access. Default: false

Return Value

The bucket
ID

Remarks

The access rights for the component will be set automatically

by the Cloud Storage component in
the ACL.


Request Message Example:


<?xml version=
"1.0"
>

<tns:cloudRequest

xmlns:tns=


http://example.com/CloudStorage.xsd


xmlns:xsi=
"http://www.w3.org/2001/XMLSchema
-
instance"

xsi:schemaLocation=
"http://
example.com
/CloudSt
orage.xsd CloudStorage.xsd "
>


<tns:createBucket>


<tns:bucketType>

Binary
</tns:bucketType>


<
tns:publicRead>false</tns:publicRead>


<tns:publicWrite>false</tns:publicWrite>


</tns:createBucket>

</tns:cloudRequest>


Response Message Example:


<?xml version=
"1.0"
>

<tns:cloudResponse

xmlns:tns=


http://example.com/CloudStorage.xsd


xmlns:xsi=
"http://www.w3.org/2001/XMLSchema
-
instance"

xsi:schemaLocation=
"http://
example.com
/CloudSt
orage.xsd CloudStorage.xsd "
>


<tns:createBucket>


<tns:success>true</tns:success>


<tns:bucketId>BUCKET1234</tns:bucketId>


</tns:createBucket>

</
tns:cloudResponse>


5.

C
ONCLUSIONS


In this paper, a new solution for using Cloud Storage in Virtual Enterprises is introduced. This
Cloud
VFIS will
enhance the conceptual framework that virtual factory use to make decisions about their manufacturing proc
ess. More
important, it will contribute to fill the gap in knowledge about the use of Cloud storage in virtual enterprise. There is an
acknowledged lack of studies about adoption and implementation processes for virtual enterprise platforms in SMEs.
Conseq
uently, we provide a concert tool to support the Virtual factory activities.


Flexible Automation and Intelligent Manufacturing, FAIM2013


In this CloudVFIS, t
he

three tier
architecture offers many advantages
to virtual factory. It is
easily expandable
and
has a

high degree of scalability.

Two additional layer load balancer and message routing complete the architecture and
it supports virtual factory enviroment.
Components

in the business layer

that need a high amount of processing power
in short time can be virtually duplicated easily

achi
ev
ed by

load balancing. The use of cloud
-
based data storage also
implies a high scalability in terms of data throughput and storage capacity. This
CloudVFIS
rel
ies

on a number of
cloud storage providers to ensure a highly elastic data
storage.
In this solu
tion, the data is managed based on the
different data types.
Moreover, t
he communication of the components

and these three tiers

will be
routed via

the
Message Routing
service.


Cloud computing is changing the way to do business in many industries. However
, this paper mainly focuses on
cloud storage implementation.

W
ith

earlier adoption of cloud storage to access/manage data will help enterprise stay
ahead, and it is very crucial to the small and medium size enterprises’ existence.

A
CKNOWLEDGEMENTS

The aut
hors would like to acknowledge the co
-
funding of the European Commission in NMP priority of the
Seventh

RTD Framework Pro
gramme (2007
-
13) for the ADVENTURE project (Adaptive Virtual Enterprise
Manufacturing Environment), Ref. 285220. The authors also ackno
wledge the valuable collaboration provided by the
project team during the research work.

R
EFERENCES


[1] J. Browne and J. Zhang, “Extended and virtual enterprises similarities and differences”, International Journal of Agile
Management Systems, Vol. 1, No.

1, pp. 30
-
36, 1999.

[2] A. Goel, H. Schmidt, and D. Gilbert, 2009, “ Towards formalizing virtual enterprise architecture”, In Enterprise Distribu
ted
Object Computing Conference Workshops, 2009. EDOCW 2009. 13th, pp. 238
-
242.

[3] J. Fredriksson and K. Augu
stsson, 2011. “Cloud Service Analysis
-
Choosing between an on
-
premise resource and a cloud
computing service”, Chalmers University of Technolohy, Göteborg, Sweden.

[4] J. Hwang et al., 2011, “A Business Model for Cloud Computing Based on a Separate Encrypt
ion and Decryption Service”,
2011 International Conference on Information Science and Applications (ICISA).

[5] H. Helaakoski, P. Iskanius, and I. Peltomaa, “Agent
-
based architecture for virtual enterprises to support agility”, Establishing
the Foundation

of Collaborative Networks: IFIP TC 5 Working Group 5.5 Eighth IFIP Working Conference on Virtual
Enterprises, pp. 299
-
305, Guimaraes, Portugal, 2007.

[6] L.M. Camarinha
-
Matos and H. Afsarmanesh, “Virtual Enterprise Modeling and Support Infrastructures: Ap
plying Multi
-
agent
System Approaches”,
Multi
-
Agent Systems and Applications, Lecture Notes in Computer Science
, vol. 2086, pp. 335
-
364,
2006.

[7] D. Chen, H. Huang and C. Ji
, “Research on Virtual Enterprise Workflow Modeling and Management System Implementation”,
WiCOM '08. 4th International Conference: Wireless Communications, Networking and Mobile Computing, 2008.


[8] Y. Zhang and M. Shi, “A Method Describing the Cooperat
ive Relationships Evolution of Virtual Enterprise”, The 8
th

International Conference on Computer Supported Cooperative Work in Design Proceedings, 2003.

[9] M.T. Martinez, et al., 2001, “Virtual Enterprise
-

Organisation, Evolution and Control”, Internation
al Journal of Production
Econmocis, vol. 74, pp.225
-
238.

[10] S.B. Yoo and Y. Kim, “Web
-
based knowledge management for sharing product data in virtual enterprises”, International
Journal of Production Economics, Vol. 75, pp. 173
-
183, 2002.

[11] S. Bergamas
chi, G. Gelati, F. Guerra and M. Vincini, “An Intelligent Data Integration Approach For Collaborative Project
Management In Virtual Enterprises”, Springer Science + Business Media, Inc. Manufactured In The Netherlands, 2005.


[12]

L.M. Kaufman, “Data Secur
ity in the World of Cloud Computing”, IEEE Security & Privacy, 2009.

[13]

P. Mell and T. Grance, 2009, “A NIST Notional Definition of Cloud Computing”,The National Institute of Standards and
Technology. http://csrc.nist.gov/groups/SNS/cloud
-
computing/cloud
-
def
-
v15.doc, [Last access: 2010
-
11
-
28].

[14]

Y. Huo et al., “A Cloud Storage Architecture Model for Data
-
Intensive Applications”, International Conference on Computer
and Management (CAMAN), 2011.

[15]


M. Hardwick , D. L. Spooner, T. Rando and K. C. Mo
rris, “Sharing manufacturing information in virtual enterprises”,
Communications of the ACM, Vol. 39, No. 2, pp. 46
-
54.

[16]

D.Petcu et al, 2012,


Portable Cloud applications

From theory to practice”. Future Generation Computer Systems,
Available online 27

January 2012.