24-10-2016, 02:05 PM
Integration of hybrid wireless networks in cloud services oriented enterprise information systems
1460804806-ContentServer1.pdf (Size: 1.75 MB / Downloads: 13)
This article presents a hybrid wireless network integration scheme in cloud
services-based enterprise information systems (EISs). With the emerging hybrid
wireless networks and cloud computing technologies, it is necessary to develop a
scheme that can seamlessly integrate these new technologies into existing EISs. By
combining the hybrid wireless networks and computing in EIS, a new framework
is proposed, which includes frontend layer, middle layer and backend layers
connected to IP EISs. Based on a collaborative architecture, cloud services
management framework and process diagram are presented. As a key feature, the
proposed approach integrates access control functionalities within the hybrid
framework that provide users with filtered views on available cloud services based
on cloud service access requirements and user security credentials. In future work,
we will implement the proposed framework over SwanMesh platform by
integrating the UPnP standard into an enterprise information system.
Introduction
In industrial sectors, it has been recognised that wireless technologies can offer signi-
ficant cost reductions as well as enabling the entirely new business processes that will not
only be less expensive, but also safer, more reliable and far more transparent than their
current ones (James 2007). The emerging hybrid wireless networks such as wireless
LAN (WLAN) (Guinard et al. 2010), wireless mesh networks (WMNs) (Zhang et al.
2009), wireless sensor networks (WSNs) (Cheng et al. 2007, Samaras et al. 2009),
Bluetooth-based Piconet, RFID-based information systems (Kamoun 2009), or mixture
of wireless-based networks make it possible to use wireless on a broad scale in manufacturing
processes, and offer opportunities to enterprises for using wireless to gain
visibility into hidden processes and activities, although such technologies have not been
well integrated into enterprises yet (Mihai 2007, Xu 2007, Xu 2008, Marstona 2011).
Over the last 10 years, more and more enterprises worldwide have adopted
enterprise information system (EIS) for running their businesses (Ralph et al. 2011). EIS provides a technology platform that enables organisations to integrate and
coordinate their business processes and information systems. In recent years, the
emerging technologies such as cloud computing, Web 2.0, WSN, RFID, or other
cutting-edge technologies make it a necessity to develop a suitable integrated EIS
(IEIS) that can integrate hybrid wireless networks and cloud computing-based
technologies for extended enterprises in a cloud environment, with the objective
of achieving efficiency, competency and competitiveness (Marstona et al. 2011).
On the other hand, hybrid wireless networks-based information exchange in
enterprises is moving towards mainstreams, which is driven largely by new
emerging wireless networks based-technologies and mobile business models
(James 2007).
With the application of wireless networks in enterprises, more and more wireless
communication techniques have been employed in enterprise information systems
(EISs), which cause the EISs more complicated and lower interoperability. It is
difficult to establish high-level interoperability and integration among different
wireless network technologies in an EIS environment, although it is of significant
importance (Samaras et al. 2009). Advances in wireless networking technologies, and
cloud computing have opened up new opportunities in EIS. Meanwhile, many
technological challenges still need to be faced. Driven by the proliferation of new
advances in technologies, easily deployed hybrid wireless networks target a large
number of applications, ranging from smart spaces, industrial processes (Chen and
Carothers 2007), to space exploration (Xu et al. 2008). Hybrid wireless networks are
very suitable for EISs due to its important features such as low cost, low complexity,
scalability, reliability and bi-directional. Hybrid wireless networks can grow easily as
the size of enterprise expands.
Meanwhile, the emergence of cloud computing represents a fundamental change
in the way EIS services are invented, developed, deployed, scaled, updated,
maintained and paid for (Xu 2007). Cloud computing promises to deliver all the
functionality of existing information services even as it dramatically reduces the
upfront costs of computing that deter many organisations from deploying many
cutting-edge EIS services (Xu 2007). In EISs, there are two main trends for cloud
computing: (a) information processing efficiency, whereby the power of computing
resources are utilised more efficiently through highly scalable resources and (b)
business agility, whereby EISs can be used as a competitive tool through rapid
deployment, parallel batch processing, use of compute-intensive business analytic
and mobile interactive applications that respond in real-time to user requirements
(Xu 2007). Cloud services can significantly lower the cost of entry for small
enterprises trying to benefit from compute-intensive business analytics. It can
provide an almost immediate access to hardware resources. Cloud computing can
also make it easier for enterprises to scale their services, which are increasingly
reliant on accurate information, according to client demand. By using the emerging
wireless network technologies, cloud computing makes it possible for new classes of
applications and delivers services that were not possible before.
This article is organised as follows: Section 2 analyses the reasons that integrate
hybrid wireless networks into EISs; a cloud-based collaborative hybrid wireless
networks framework is proposed in Section 3; Section 4 discusses the serviceoriented
management in IEIS. The cloud data structures in IEIS are given in Section
5, and the security strategies for the proposed architecture are proposed in Section 6.
Finally, Section 7 concludes the article.
The need for integrated EIS with wireless hybrid wireless networks
More and more enterprises have adopted information systems for running their
business, which includes Enterprise Resource Planning (ERP), supplier relationship
management (SRM), customer relationship management (CRM), CAD, CAM, or
their integration within a business organisation. With global operation, global supply
chain, and fierce competition in place, there is a need for suitable EIS such as ERP,
E-Business or E-Commerce systems to integrate extended enterprises in a supply
chain environment with the objective of achieving efficiency, competency, and
competitiveness (Marstona et al. 2011). The application-oriented EISs have been
widely used in enterprise’s business models. The hybrid wireless networks and cloud
computing also make new classes of applications possible that delivers services
through EISs that were not possible before, such as mobile interactive applications,
parallel batch processing applications, and business analytics that can use the vast
amount of computer resources to understand customers, and the extension of
compute-intensive desktop applications.
In this section, we will discuss the need of IEIS from four aspects.
2.1. Scalability and robustness of enterprise information systems
The IP network -based EISs can be effectively developed for different enterprises.
However, the scalability and robustness of the existing EISs may cause low
efficiency for new wireless technologies to be applied. Hybrid wireless networks
may bridge the gap between existing EISs and the new cloud based applications.
On the other hand, hybrid wireless networks can easily, effectively, and wirelessly
connect the different components in EISs, such as data centres, CAD, CAM, and
E-Commerce systems that using different wireless communication technologies.
Since these wireless networks are complementary rather than competing with
each other, it is possible that these different wireless communication technologies
are used in the same place to form hybrid wireless networks over the same
spectrum band.
2.2. Collaborative information exchange for peer-to-peer users over wireless EIS
In EISs, the peer-to-peer collaboration paradigm fundamentally changes the
current passive manner through which the wireless networks adapt their
transmission strategies to matching available resources, enabling them to influence
system dynamics proactively through exchange of information and resources.
The integrated EIS (IEIS) with hybrid wireless networks can effectively connect
the different information systems in large enterprises. It can help the enterprises
to develop a distributed and efficient framework for resource exchanges that
enables peers to collaboratively distribute available resources among themselves
based on the quality of service requirements. Hybrid wireless networks based
collaborative approaches can achieve improvement over non-collaborative
approaches.
2.3. Cooperation for effectiveness in wireless IEIS
In a hybrid wireless networks IEIS, the components should be able to work
properly without interfering with each other. One concern is the coexistence of different devices in IEIS. It is clear that traditional devices are deployed without
interference; however, the wireless devices used in hybrid wireless networks may
interfere with each other. On the other hand, the factors such as communication
protocols, and information format in transmission should be taken into
consideration.
IEIS can offer a more cost-effective means of monitoring plant equipment and
production processes, and enables real-time decision making to optimise production
or to eliminate maintenance issues before they interrupt production. Literally
millions of field devices are installed at dear cost in process manufacturing facilities.
However, because most are not digitally enabled, their ability to share process and
maintenance information is extremely limited. This presents a huge opportunity for
wireless technology, which can be used to enable these stranded assets to the benefit
of operations, maintenance, and business systems across the enterprises.
2.4. Network characteristics of IEIS
Large-scale data centres, particularly those with a combination of physical
infrastructure and a proliferation of virtual servers require new approaches to
deliver and manage information system resources. For the enterprises, it is important
to improve the efficiency of resources management. The IEIS should be able to
effectively manage and utilise the assets over a shorter period of time. On the other
hand, the current hardware assets that are over four years old should be replaced
prior to starting implementation of the new EIS model.
The older assets are generally underperforming and typically require more
maintenance than the latest offering (Ralph et al. 2011). IEIS particularly in the
infrastructure aspect can enable enterprises to deliver services quickly and manage
services efficiently. The data storage servers and networks will work very closely
together and in some cases will be part of the same device. The IEIS may offer benefits
such as simplified management, increased availability, flexibility, higher utilisation,
and higher resilience inside the infrastructure, due to the fact that the IEIS
infrastructure can scale on demand in a way that is safe and consistent (Sinderen and
Almedia 2011).
2.4.1. Coexistence
In our previous work, we have developed a compressed sensing based existence
mechanism for hybrid wireless networks. This technique makes it possible for devices
in hybrid wireless sensor networks to work together with low packet loss (caused
by interference). By this way, short-range wireless communication systems like
Bluetooth, ZigBee and IEEE 802.11 (WiFi) that are the most commonly deployed
technologies for WPAN, WSN and WLAN can coexist in EISs with low interference
(Kakousis et al. 2011). The hybrid wireless networks in the whole systems assume a
gateway role (Li and Wang 2011). They have to handle the communication with
devices, sensor networks, RFID system, and other information collectors. Since the
resource constraints of the WSNs, the host services offered by WSNs are in the form
of web services, they can be accessed by the backend systems. In order to facilitate
the coverage of gateways over WSNs, different gateway mechanisms are used that
are equipped with different wireless communication technologies such as Zigbee
(IEEE 802.15.4), Bluetooth, or WiFi. Furthermore, the WSNs can be accessed via
168 S. Li et al.
GPRS, and the gateways will be using a lightweight service discovery protocol
(So and Vaidya 2006).
2.4.2. Protocols
In hybrid wireless networks, many types of communication protocols are applied. In
our model, the system adaptively selects the protocols depending on the wireless
network types. Since in WMN the network traffic should be higher than that in
WSN, we use the load balanced multipath protocols to balance the traffic. Actually
many effective multipath protocols have been developed by Mateljan et al. (2010)
and Kakousis et al. (2011) that can be directly used in IEISs.
2.4.3. Information format
Our proposed approach pursues the integration of hybrid wireless network into EIS,
particularly for services-oriented applications, the information collected by the
frontend layer should be recognised and employed by entities in other layers such as
backend layer, or middleware layers. So it is necessary to develop commonly
recognised information format. In this paper, we use the Web Services Description
Language (WSDL) to describe the services description, which can effectively share the
services between layers. These services can be hosted on different devices, e.g. sensor
nodes could potentially call services running on more powerful devices outside the
sensor networks (Beheshti and Beheshti 2010, Suresh and Rajendra 2007).
3. Cloud-based collaborative hybrid wireless networks framework in enterprise
information systems
It has been confessed that many cloud-based applications today lack some of the
functionality of their traditional counterparts (Xu 2007). Especially for the complex
EISs, the applications might not be temporarily suitable for transition to a cloud but
might nevertheless need to interact with other cloud-based applications. Cloud
applications are not yet equipped with the availability or QoS that meet some
organisational requirements from their EIS systems (Iacob and Jonkers 2009,
Cucinotta et al. 2010). Similar to other centrally located services, cloud services are
subject to outages or even data loss that could result from various reasons as
hardware and/or software failure to acts of nature.
Typically, deployment of an IEIS includes three phases (Cucinotta 2010, Xu
2011, Zhou et al. 2011):
(1) Integration of traditional hardware functionality, such as combining servers,
data processors, hybrid wireless networks, and existing infrastructures;
(2) Tightly integrate information management system (such as ERP, CRM,
SRM, CAD/CAM, etc.) and control software into hardware functionality;
(3) Deploying an IEIS requires integrating various EIS services such as
outsourcing, software as a service (SaaS), platform as a service (PaaS), and
cloud computing into a unified computing framework.
One of advantages of IEIS is that all the physical elements are virtualised,
whether they are computing or storage resources, which do not have to reside in a physical building or even be a part of a physical process. Therefore, an IEIS can lead
to simplified management, increased flexibility, better availability, and higher
utilisation.
Figure 1 shows the cloud architecture of an IEIS, which maps the integration of
an EIS into a private cloud. It can provide various cloud services. From Figure 1, it
can be seen that an IEIS includes three-layer architecture: the front end layer, the
middle layer, and the backend layer, which are connected with hybrid wireless
networks that involved in EIS. In the front end layer, infrastructure as a service
(IaaS) is built on top of virtualised compute storage, and network resource, platform
as a service (PaaS) at the OS/middleware level, and software as a service (SaaS) at the
user application backend layer (Hwang and Li 2010, Huang and Hsiung 2011).
The application module in an IEIS can provide services, data analysis, Web
applications, and security protection. The data/information module provides data loss protection, service log file, service activity, monitoring, and data processing.
The management provides services such as remote management interface, and
networking is able to provide network services and protection. The computing/
storage can provide cloud computing resource, cloud data storage and integrity,
encryption, and masking (Erol et al. 2010, Gaukler 2011).
3.1. Hybrid wireless network-based IEIS architecture
It is critical for IEIS to support business in a way that responds to business needs in
real-time. One of the advantages of IEIS is that there will be little or no lag between
application development and deployment once enterprise gets through the
development (Cucinotta et al. 2010). Once the applications are in a production
environment and there are swings in business demand, enterprises will be able to
gradually increase or decrease that demand in a fairly quick and seamless manner.
The ability to do this represents a tremendous cost savings because the return of
investment (ROI) for those applications and the infrastructures now becomes very
short. Over an IEIS platform, an enterprise can deliver a standard package with
layer virtualisation and systems management software in it. In this way, an enterprise
does not have to manage more and more devices, instead, can provide more valueoriented
functions as cloud services (Ulusoy et al. 2011).
In this section, we define a three-layer service-oriented architecture to integrate
the hybrid wireless networks into EISs, which is able to bridge the new coming
wireless technologies in existing networks. In Figure 2, the three layers as the
backend (based on existing EIS systems), the middleware networks (interface
description, information exchange between mini-database to the database), and the
front-end (include the infrastructure and the maintain networks, such as WSN,
RFID, etc.) are illustrated and more detailed discussion will be given in the following
subsections (Ramamurthy et al. 2007, Estevez Ayres et al. 2009).
3.1.1. The backend layer
This layer is responsible for the business applications which can access the services
offered by the EIS at the level of web services. As shown in Figure 2, the backend
layer includes the existing EIS that can be implemented completely independent of
the underling platforms.
The backend layer contains five high-level components: (a) Service repository – it
contains a database of the available services including their description and
implementation. A single service may have several implementations on different
platforms. The server descriptions are XML documents formatted in CoBIL
(AbdelSalam et al. 2010). The interface definition is derived from Web Service
Description Language (WSDL); (b) System State Manager, which stores the
operational state of the components in IEIS; © Service Mapper – it provides the
systematic mappings of the services to the implementing node according to the
services description; (d) Service Scheduler – it processes the service schedule
according to the requirements that issued by the application running on the backend;
and (e) Cloud Service Inventory – it implements a web service interface for
cloud services distribution over the public cloud. The user applications can register
with the cloud services inventory manager and receive notification when the services are available.
This layer is responsible for cloud publishing, Web services designing, and the
communication framework for supporting interoperable interaction over a network.
The communication between the Web services server and client is through Simple
Object Access Protocol (SOAP), which sends and receives XML message via
HTTPS. This layer often provides a machine-readable description of the operations
written in the WSDL, and the client obtains the type and function information by
acquiring the service description by the WSDL from the server side (Chiang et al.
2011).
3.1.2. The middle layer
We use WMN to implement the middle layer. The reason is that WMN can provide
reliable, flexible, and effective networking which supports the UPnP standard as the
uniform interface between the application layer and the frontend layer. This layer
provides the following functionalities: (a) Message transformation, it handles the
packet-level translation between the proprietary messages and UPnP arguments; (b)
Service Manager assists the deployment of new services in the network; © Service
proxies are responsible for cloud services deployment. Deployment requests are
Figure 2. Logical architecture of service-oriented integrated enterprise information system.
172 S. Li et al.
issued to the gateway by specifying the executable service and the XML-based
deployment description.
One of the key features of this layer is the dynamic entity of service proxies,
which can be accessed as native UPnP devices. In this layer, the WMN is
characterised by four features:
(a) Security. High-performance operating system was built from the ground up,
which is robust and capable of auto-configuring with minimal administrative
intervention.
(b) Over-the-Air (OTA) Firmware Upgrades. Stay up-to-date through upgrades
that occur in the background, so that networks continue operate with
minimal downtime.
© Controller. Secure network tunnels provide highly available networking
configuration, statistics, and monitoring without any special configuration.
(d) UPnP supports mesh networks, which simplify the installation and the
administration of network devices and services, enabling them to control and
to be controlled, to discover and to be discovered. UPnP can be easily
implemented on mesh networks. UPnP supported device may provide
different sets of services.
3.1.3. The front-end layer
In enterprises many infrastructure devices and wireless networks are involved,
such as WSNs, RFID systems, EIS diagnose networks, mobile networks, industrial
monitoring networks, etc. It can vary depending on the applications, in which the
information format is needed for enabling effective integration, access, fusion, and
use of sensor-derived data in applications. Distributed processing enables scalable
performance using limited resources and guarantees gracious degradation under
network failures. Redundancy in processing capability ensures that the hybrid
wireless networks are more or less operational throughout the process but this
assurance comes with additional challenges in network management, flow of
information, and task allocation.
3.2. Cloud based services in enterprise information systems
Cloud computing is quickly becoming one of the most significant technologies of our
times. Many enterprise level cloud technologies have been developed such as grid
computing, virtualisation, etc. The cloud computing can be defined as: ‘An emerging
computing paradigm where data and services reside in massively scalable data
centres and can be ubiquitously accessed from any connected devices over the
Internet’ (Xu 2007). Cloud Computing combined with process-centric Platforms-asa-Service
offers a unique opportunity to develop a new class of enterprise systems
(Suresh and Rajendra 2007, Zhou et al. 2011).
In an IEIS, the business units can negotiate with EIS for the level of service in a
genuine fashion. The main trend in the market is growth through incorporation of
new wireless technologies. In this section, our goal is to propose an IEIS which can
effectively integrate the existing IP-based EIS and new wireless technologies and
provide cloud services. The IEIS is able to provide following features (Hwang and Li
2010, Wang et al. 2011, Zhou et al. 2011):
Enterprise Information Systems 173
. SaaS. Delivering pieces of an application on demand as a service to a customer.
Underneath SaaS there are two basic components: (1) infrastructure as a
service (IaaS), which maintains the hardware, server, storage as services; and
(2) platform as a service (PaaS), in which the requirements for platform are
delivered as a service.
. As the infrastructure changes, the IEIS needs to have a better understanding of
its cost structure to improve benefits to the business. IEIS has to change its
chargeback processes to focus on services rather than assets, which require
transparency and visibility into what end users are consuming.
. An enterprise can provide transparency by allowing IEIS to configure and
provide services for the business. By doing this, the enterprise has the
opportunity to dramatically change its chargeback mechanisms from asset
based to service based, using a billing process for different types of services.
. An IEIS includes integrating work, storages and server components, which
should be capable of supporting many users. Organisations with existing
infrastructure may also want to deploy IEIS infrastructure technology and
components from vendors that maintain a large ecosystem of partners
comprising global systems integrators, service providers, and channel partners.
. IEIS should provide enterprise information system management across a
unified infrastructure of storage, computing, networks, and virtualisation
resources to allow customers to accelerate their transition from a physical data
centre to a virtualised data centre, and then to private service clouds.
. IEIS will allow customers to manage their infrastructure in a unified manner.
Therefore the IEIS can provide interfaces to manage and automatically
configure, discover, provision, and monitor across the infrastructures, including
networks, servers, and storage centres as well as virtualised applications.
. Cloud services should be managed effectively in an IEIS, and more details
about cloud service deployment will be discussed in Section 4.
Figure 3 illustrates the cloud architecture for the proposed model.
3.3. Multi-enterprise information systems
While steadily becoming a paramount viable choice, SaaS has led the way to a variety
of services offerings accessible through the Web services (Marstona et al. 2011): from
infrastructure to business information, supply chain, and enterprise partner
management. Multi-Enterprise Information Systems (MEISs) are about providing
these services more effectively and valuable to complicated business transactions and
can be suitable for any businesses, large or small (Gaukler 2011). The basic objective
of an MEIS is to quickly enact a partnership that is able to improve visibility,
managing exceptions, facilitating information and data exchange. MEIS helps
enterprise to improve the efficiency of infrastructures used and assist with compliance
or accelerate joint-ventures or mergers. MEIS can be implemented by collaborating
on particular activities between enterprises information systems.
MEIS may lead to significant inefficiencies, risks, and inflexibility. Since one of
the parties may have to bear all the costs of MEIS, a well-built MEIS can
significantly simplify the business process. Actually, it has been proven that, for the
portal of MEIS, it is difficult to integrate with the party systems. The emerging cloud
computing makes it possible to deliver MEISs where each party can contribute to the
174 S. Li et al.
development of the entire system. By cloud computing, the development costs can be
significantly reduced and it can provide superior integration capabilities with all
party’s systems and web-based intranets applications.
Figure 4 represents the elements of an MEISs built on a Platform-as-a-Service
foundation, in which the interfaces can be used to access services, by applying any
combination of web based technology (SOAP, ReST, ftp, http, etc.) and format
(EDI, XML, etc.) (Ramamurthy et al. 2007, Estevez Ayres 2009, Gaukler 2011).
MEIS also integrates the human workflows, in which the process can be constructed
iteratively.
Enterprise collaboration on-demand information exchange between services cloud
In an IEIS model, the on-demand information exchange between services makes it
possible to collaboratively provide complicated services to the applications. The ondemand
information exchange between services can be implemented by metadata
technology, which works on the platform of thin computing, e.g. sensor networks,
RFID chips, thin infrastructures, as well as traditional databases. In our model, a
global services module is proposed that can support participants to offer/publish
information according to the request/subscription and the information that fits.
4. Service-oriented multiple-enterprise information system management
In this section, we will discuss an overall services management architecture based on
the IEIS model. The benefits that can be reached by utilising service-oriented IEIS
are numerous; hence, there is a need to move towards such approaches. The most
prominent advantages that can be obtained by IEIS include reducing cost, and
improving flexibility and agility to respond to dynamic business conditions.
The mismanaged services of IEIS users can lead to the fault implementation
compromising the integrity of the services ecosystem and will become very difficult to
manage the services in the cloud service environment. In this section, we will provide
management scheme for services in cloud computing environment. As the number of
technological abstractions continues to grow, it will become increasingly difficult to
manage highly virtualised cloud services. In order to deal with this growing
technological complexity, intelligent and real-time service management scheme will
be needed for growing cloud computing environments for self-managing, discovering
and analysing the cloud services.
In the service-oriented architecture definition process or during the business
process modelling, the services tag of the providers and consumers becomes very
important for performing the handshaking and meeting expectation via the service
level agreements. In Figure 5, a novel services management framework is proposed,
which consists of the following components:
(1) Enterprise services repository – it is a Meta repository that contains the
information of the enterprise services. Repository can be single instance or
multi-instances, in which the cloud services for consumer should be stored in
the single repository and then distributed to the physical directories.