A Multi-Service Multi-role Integrated Information Model for Dynamic Resource Discovery in Virtual Networks

May El Barachi1, Sleiman Rabah2, Nadjia Kara3, Rachida Dssouli2, and Joey Paquet2

1 Zayed University, Khalifa City B, P.O. Box 144534, Abu Dhabi, United Arab Emirates 2Concordia University, 1455 De Maisonneuve, Montreal, Quebec, H3G 1M8, Canada

3 École de Technologie Supérieure, University of Quebec, Montreal, Quebec, H3C 1K3, Canada Abstract --- Network virtualization is considered as a promising

way to overcome the limitations and fight the gradual ossification

of the current Internet infrastructure. The network virtualization

concept consists in the dynamic creation of several co-existing

logical network instances (or virtual networks) over a shared

physical network infrastructure. One of the challenges associated

with this concept is the dynamic discovery and selection of virtual

resources that can be composed to form virtual networks. To

achieve that task, there is a need for a formal and expressive

information model facilitating information representation and

sharing between the various roles/entities involved. We have

previously proposed a service-oriented hierarchical business

model for virtual networking environments, as well as an

architecture enabling its realization. In this paper, we build on

this business model and architecture by proposing a multi-

service, multi-role hierarchical information model, for virtual

networking environments. Furthermore, we demonstrate the

usage of this information model using a secure content

distribution scenario that is realized using REST interfaces.

Unlike other proposals, our integrated information model enables

the fine-grained description of virtual networks and virtual

networking resources, in addition to the modeling of network

services and roles, and their relationships and hierarchy.

Keywords - Network virtualization, dynamic resource discovery, information modeling, secure content distribution, REST.

I. INTRODUCTION

The concept of virtualization consists in the decoupling of physical resources from the service-level view, by adding an abstract layer (software), in between. The implementation of this concept gives the end-user the illusion of direct interaction with the physical resources, while allowing efficient utilization of resources/infrastructures and enhanced flexibility. Different forms of virtualization have been proposed, such as: Storage virtualization referring to the separation of physical disk space from the logical assignment of that space using various techniques (e.g. RAID and SAN); server virtualization consisting in the partitioning of the resources of a single physical machine into multiple execution environments (or virtual machines), each running its own operation system and server applications; and application virtualization referring to the isolation of a certain application from the operating system on which it runs as means to achieve OS-independence. Network virtualization is an emerging concept that applies

virtualization to entire networks. The basic idea behind network virtualization consists in the dynamic creation of several co-existing logical network instances (or virtual networks) over a shared physical network infrastructure [1].

Unlike Virtual Private Networks (VPNs) [2] that are limited to traffic isolation capabilities and do not allow customization nor administrative control, virtual networks (VNets) can be built according to different design criteria and operated as service tailored networks. Network virtualization is considered as a promising way to

overcome the limitations and fight the gradual ossification of the current Internet. Beyond the Internet’s context, there are several important motivations behind the network virtualization concept. One of these motivations is the cost effective sharing of physical networking resources, by partitioning the resources of an existing infrastructure into slices and the allocation of these slices to different VNets (operated by different service providers). Another motivation is the potential for having customizable and service tailored networking solutions via the addition of new technologies or customized versions of existing technologies, in the virtualization layer. We have previously proposed a service-oriented hierarchical

business model for virtual networking environments [3], as well as an architecture enabling its realization [4]. In this model and its related architecture, different levels of services (i.e. essential services, service enablers, service building blocks, and end-user services) offered by various roles, can be dynamically discovered, used, and composed. To enable the dynamic discovery and selection of virtual

resources that can be composed to form virtual networks, there is a need for an expressive information model facilitating information representation and sharing. This model should support the description of the functional and non-functional aspects of available resources and services, while enabling the interaction between various roles in virtual networking environments. In this paper, we propose such a model and illustrate its utilization using a concrete scenario. Unlike other proposals, our model is an integrated model enabling the representation of physical/virtual resources and services, as well as the mapping between them. Furthermore, it models the relations between multiple business roles and their association to resources and services offered. The next section gives an overview of the previously

proposed virtual networking business model and architecture. This is followed by an elaboration of our integrated information model, and a REST-based content distribution scenario, illustrating the model’s utilization for dynamic resource discovery and selection. We end the paper with a discussion of related work, before drawing our conclusions.

978-1-4673-5939-9/13/$31.00 ©2013 IEEE978-1-4673-5939-9/13/$31.00 ©2013 IEEE

2013 IEEE Wireless Communications and Networking Conference (WCNC): SERVICES & APPLICATIONS2013 IEEE Wireless Communications and Networking Conference (WCNC): SERVICES & APPLICATIONS

4777

II. OVERVIEW OF OPEN VIRTUAL PLAYGROUND ARCHITECTURE & RELATED BUSINESS MODEL

This section introduces our previously proposed business model for virtual networking environments, and gives an overview of the Open Virtual Playground architecture we proposed to realize this business model.

2.1 Proposed virtual networking business model

Figure 1 depicts the business model we proposed in [3] for virtualized networking environments. Four levels of service are defined as part of our model, namely: Essential services constituting mandatory services needed for the basic operation of the network (i.e. routing/transport services); Service

enablers consisting of the common functions needed to support the operation of end-user services (e.g. session/subscription management, charging, security, and QoS management); Service building blocks acting as elementary services that can be used/combined to form more complex services (e.g. presence and call control); and End user services constituting the value-added services offered to users.

Figure 1. Proposed business model for virtual networking environment As shown in this figure, our business model defines five

business roles, namely: 1) The Physical Infrastructure

Provider (PIP) that owns and manages a physical network infrastructure and can partition its resources using some virtualization technology. The services offered by the PIP are essential bearer services. 2) The Service Provider (SP) that has a business agreement with the subscriber and offers value added services, which could be simple or composite (i.e. formed by combining service building blocks); 3) The Virtual

Infrastructure Provider (VIP) that finds and aggregates virtual resources (offered by one or more PIPs), deploys any protocols/technologies in the instantiated VNet, and operates it as a native network. The VIP supports SPs or other VIPs with service enablers and service building blocks and has no direct business agreement with consumers; 4) The Consumer who acts as the subscriber and the end user of value added services; and 5) The Services and Resources Registry (SRR) acting as broker by providing information to find other parties and the services/resources they offer.

2.2 Overview of the Open Virtual Playground architecture

The Open Virtual Playground architecture is a layered architecture that we proposed in [4], to enable the realization of our business model. This architecture introduces data and

control planes at each level of the hierarchy. While the data plane provides essential data transportation functionality, the control plane encompasses all the control and management functions needed for the provisioning of different levels of services. Three hierarchical levels are proposed in our architecture: The physical network level (managed by the PIP); the first virtual network level (managed by a VIP); and the second virtual network level (managed by a SP). At the physical network level, we find the physical data plane containing regular and virtual routers connected to form the physical network infrastructure, as well as the physical control plane responsible of the following functions: Resource publication, resource negotiation, resource allocation and provisioning, and resource management. At the first virtual network level, we find a virtual data plane encompassing a set of virtual nodes connected by virtual links (essentially a subset of the underlying physical topology), as well as virtual control plane 1. This latter encompasses the following functions: a set of service enablers and service building blocks, service publication, resource negotiation, resource discovery/selection, and service deployment/management. Similarly, the second virtual network level consists of a virtual data plane and a virtual control plane. The latter is responsible of the following: a set of end user services, end user service publication, resource negotiation, resource discovery/selection, service deployment/management, and service composition. Distributed across all the levels of the hierarchy, we find a service and resource registry, which is responsible of the formation/maintenance of a global view of the physical/virtual networks contexts and available services/resources. More details about our architecture can be found in reference [4].

III. INFORMATION MODELING AND DYNAMIC RESOURCE DISCOVERY APPROACH

Virtual networks are federated networks created by aggregating virtual resources that may span across multiple network domains. To enable the interaction between various roles in a virtual networking environment (VNE), these roles need to agree on a single and common data representation model, allowing them to efficiently and reliably exchange information (i.e. resources’ description, service description, VN query, etc.). In this section, we introduce an integrated hierarchical information model that aims at modeling virtual networks, virtual network resources, virtual network services and their hierarchy, as well as virtual network roles and their relationships.

3.1 Integrated Hierarchical Information Model

Several requirements have led to the design of our proposed information model, namely: the fine-grained description of physical/virtual resources and services (including functional and non-functional attributes and constraints); Virtual-to-

physical resources and topologies mapping support; formality

and expressiveness; multi-role communication support; network resources/services composition support; as well as platform-independence. Figure 2 shows a high level view of

Consumer (end user / subscriber)

Services & Resources

Registry (broker)

Virtual Infrastructure Provider (3rd party service provider)

Service Provider

Physical Infrastructure

Provider

4778

our integrated information model. This model is inspired by the work introduced in [5, 6].

Figure 2. High level view of proposed information model

Our information model revolves around three main concepts and their relationships: roles; services; and resources. Roles are entities that collaborate to offer/consume resources and services and exchange information related to these resources/services. A role can act as resource provider offering and managing virtualized resources, or as a resource consumer accessing virtualized resources. In addition, a role can act as service provider offering and managing network services, or as a service consumer subscribing to network services. In our model, network resources are mapped onto network services (i.e. network resources are considered as low level network services). Furthermore, roles are considered to be distributed and loosely coupled entities interacting via programmable interfaces. Finally, just like web services, various levels of network services can be published, dynamically discovered, composed, and used, in our model. In Figure 2, we model the different roles and their

relationships to physical/virtual topologies and various levels of services. We consider a TargetedNetwork to be the base entity as well as the root element of all instantiated description documents. A TargetedNetwork can be composed of one or many virtual networks and one or many physical networks. A PhysicalNetwork has a PhysicalNetworkTopology and is composed of a set of physical nodes connected by physical links. A VirtualNetwork has a VirtualNetworkTopology, which is a subset of the underlying physical topology. A virtual network topology can be composed of one or multiple virtual

ones, thus forming a vertical hierarchy. A virtual network is composed of a set of VirtualNodes, each node having one or many VirtualInterfaces and being connected to another virtual node by a VirtualLink. Virtual nodes that are instantiated on the same physical device are grouped in a VirtualNodeGroup that is mapped to a physical node. Although we are not concerned about modeling physical networks related entities, we only model a physical network as a set of PhysicalNodes where a given group of virtual nodes is mapped. The different roles and their interactions with different

entities are modeled as follows: (1) A PhysicalInfrProvider (PIP) owns and operates a PhysicalNetwork; offers EssentialServices; and instantiates one or multiple VirtualNetworks; (2) A VirtualInfrProvider (VIP) manages and operates VirtualNetworks and offers ServiceEnablers; (3) A ServiceProvider (SP) manages and operates VirtualNetworks and offers ServiceBuildingBlocks and EndUserServices. An end user service can be created by combining one or more service building block services; and (4) Considered as end-user, a Consumer subscribes to/uses one or multiple EndUserServices that are accessible via PhysicalNetworks and VirtualNetworks. To further detail our model, Figures 3 and 4 respectively

show the resource level view and the service level view.

Figure 3. Resource level view of proposed information model

4779

In the resource level view, we consider as the basic building component of a virtual network, a NetworkElement (NE) that can be a Node, Link, Interface, or Path. A NE has a name, availability, start time that specifies when the resource is available, and a period that determines for how long the resource is available. The status attributes represent NE’s state (available, allocated, etc.). A NE belongs to a NetworkDomain, which in turn has an AdministrativeDomain. A Node can be either a PhysicalNode or VirtualNode.

Represented in the class Node, a node has a GeoLocation and encompasses common attributes needed for describing a network node, namely, a network stack, a type (i.e virtual switch, virtual router, virtual machine, etc) and an IP address. Besides attributes such as the vendor, model, and substrate node group, a physical node may aggregate virtual nodes and interfaces, whereas a VirtualNode (VN) is uniquely identified; and has an initial and maximum capacity in terms of computational capabilities. Each VN aggregates one or multiple virtual interfaces. An Interface represents a physical/virtual network interface controller (NIC); and has a type (i.e Ethernet, radio), rate and MAC address. Depending on its capacity, a physical link can be divided into slices using virtualization techniques (i.e ATM, MPLS) to support one or multiple virtual links. A Link has characteristics such as minimal delay, type, bandwidth, throughput, good-put and type of connectivity; and an end point that determines the source node and destination node. Each VirtualLink has a tag, and initial and maximum allocated bandwidth. Virtual interfaces are connected by a virtual link. A physicalLink has a limited number of supported virtual links and an additional attribute for defining available bandwidth. A Path represents a set of links. A path starts at beginNode and ends at endNode. To represent nodes’ functional and non-functional

characteristics, a node has an association with the following two entities: (1) Node Functional Parameters: consists of characteristics/properties related to the functioning of a node

such as operating system type, software version, and the type of the network management system. It is composed of: (a) Storage parameters which determine the available disk space, storage type, and number of storage units; (b) memory parameters which represent the size, capacity, and type of the available memory; and (c) CPU parameters which represent the information about the available processing unit(s). (2) Node Non-Functional Parameters: this class defines constraints, QoS scheme, and desired criteria that should be met when selecting a resource, namely: cost, rank, and percentage of failure. In turn, non-functional attributes are composed of the following: (a) Performance parameters representing node performance properties such as response time, uptime, capacity, and reliability level. (b) Security level

parameters defining security properties that a node supports like hashing techniques (i.e Checksums, cryptographic hash functions), encryption methods (i.e symmetric, asymmetric) and security properties (i.e confidentiality, integrity). (c) QoS

parameters representing QoS related characteristics including the average packet loss, jitter, delay, and bit rate.

We model network topology as physical/virtual topology. In general, a network topology has name, type (i.e bus, ring), path list, and is composed of a set of nodes. Representing the topology of a virtual network, a virtual topology is a subset of a physical one and can be hierarchical so that a virtual

topology can be instantiated on top of one or multiple virtual topologies. Thus, this leads to have hierarchical associations among VNets. Besides, it contains attributes related to availability, start time, period, and a reference to its operator.

Figure 4. Service level view of proposed information model

In the service level view shown in Figure 4, a role represents an organization, identified by a name or id and has contact information. Different roles are modeled as follows: (1) broker represents the SRR; (2) Service provider represents a SP; (3) consumer represents an end user which subscribes to services offered by a SP; (4) Physical infrastructure provider

represents a PIP; (5) Virtual infrastructure provider represents a VIP. Each role is associated with a service entity which indicates the type of service he offers. Just like NE, a service represents the base class for

describing services. A service has the following sub-classes: (1) description and discovery service offered by the broker and representing services needed for publishing and discovering resources/services; (2) Essential service are transport service and routing service; (3) End-user service

representing services destined to end users and composed of one or many service building blocks namely call control, presence, conferencing, and messaging; and (4) Service

enablers defining the support functions needed for the operation of end user services. Service enablers comprise the

4780

following: Interworking, security level, session management, subscription management, AAA service, QoS control, media

handler. Each service is associated with functional attributes as well as non-functional attributes. We divide the latter into three categories: (1) QoS defining characteristics such as the offered class of service, support level, error rate, average repair time, and transmission delay; (2) Service performance representing properties that are related to service performance, namely, scalability and fault tolerance, response time, and uptime percentage, etc.; and (3) service security defining the security service and the level supported. Furthermore, common properties like service rank, cost, and maximum number of supported users can be expressed as well.

3.2 Model Usage in a Concrete Scenario

Figure 5 illustrates the usage of our proposed information model for dynamic resource discovery and selection, in a secure content distribution scenario.

Figure 5. REST-based virtual networking content distribution scenario

In this scenario, we find the following roles: a PIP managing the infrastructure offering communication capabilities; a VIP instantiating VN1 to offer security, QoS, and content-based routing as service enablers; and a SP instantiating VN2 to offer the secure content distribution value added service to consumers. To realize this scenario, we used REST-based interactions between the different entities. REST [7] is a network architectural style for distributed hypermedia systems. It has been gaining a lot of popularity for its use of basic web

technologies for building and provisioning distributed services. There are several motivations behind our choice of REST-based interfaces, namely: its reliance on existing well known standards opens the door for various players to adopt such unified interfaces, for which the necessary infrastructure has already become pervasive; the fact that REST is simple, lightweight, and easy to develop; and the fact that it is resource-oriented and supports a wide range of resource description mechanisms and representation formats. As shown in the figure, the scenario starts when a PIP

publishes a description of the resources (step 1) it offers as well as their related constraints in a document that is used to populate the broker, using a POST request. In this request, the broker’s resource creation service URI is specified. Once the resources’ descriptions are created, a 200 OK message (step 2) is sent back to the PIP. In turn, the VIP (wishing to create VN1) sends a PIP discovery request (step 3) containing a document describing the resources to be leased, their desired availability, cost, and constraints. This request is sent using a GET message to the broker, which replies back (step 4) with a list of available providers that can satisfy the specified requirements. Upon receiving the PIPs list, the best PIP is selected by the VIP, using a selection/matching algorithm (step 5). The VIP then sends a resource negotiation request (step 6), specifying the requested essential services and their constraints, to the selected PIP. The latter replies with a resource negotiation response (step 7), specifying the offered resources and accepted constraints to the VIP, which concludes the negotiation process with a resource negotiation acknowledgement (step 8) confirming the negotiated resources and constraints. At this stage, the PIP carries a resource allocation and virtual topology instantiation process for VN1 (step 9), and sends an acknowledgement (step 10) of the topology instantiation to the VIP. Afterwards, the VIP asks the PIP to deploy and test the specified service enablers (step 11), and gets a 200 OK message as reply (step 12). Once the service enablers are deployed and tested, the VIP asks the broker to publish a description of the service enablers and their constraints (step 13), which in case of success results in a 200 OK message (step 14). Meanwhile, a SP (wishing to create VN2) sends the broker a

VIP discovery request (step 15) containing a document describing the service enablers to be used, their desired availability, cost, and constraints. The broker replies with a list of VIPs offering service enablers that comply with the request (step 16). Later, in step 17, the SP selects the best VIP to which he submits a service negotiation request (step 18). In steps 19 to 28, interactions related to service enablers’ usage negotiation, VN2 topology instantiation, and the deployment of the content distribution end user service offered by the SP are carried, similarly to the VIP::VN1 case. The main difference lays in the message parameters that refer to a different type of service in this case. When the end user service is successfully deployed and tested, the SP sends its description to the broker. This description is then discovered

4781

by the consumer that uses it to select the best SP. Afterwards, the consumer submits a bind and invoke service request to the chosen SP, which in response sends an acknowledgment and grants access to the consumer. The latter then carries the rest of the interactions related to the end user service invocation and usage (those interactions are not shown in the figure).

IV. RELATED WORK

Several resource description languages and specifications have been proposed in areas such as grid computing, cloud computing, and the Semantic Web, while few other models were proposed for virtual networking environments. In this section, we discuss the solutions that are the most relevant to our work. Based on the Resource Description Framework (RDF) [8]

and Semantic Web, the Network Description Language (NDL) [9] has been mainly introduced to model hybrid networks using a modular information model encompassing five independent schemas, namely: topology, layer, capability, domain, and physical schema. Although NDL supports various networking aspects and has been extended in many works, it lacks support for virtual networking concepts and does not allow the specification of constraints and QoS parameters. Considering virtual resources as services with minimum

granularity, authors in [10] propose a WSDL-based model for virtual network resource description. The main goal is to support the dynamic update of resources’ information. In this model, each resource description defines the nature of the resources being offered and the location where these resources can be accessed. This solution does not provide support for detailed virtual resources nor network services modeling. The Flexible Resource Description Language (FLeRD) [11]

is proposed for virtual networks connecting cloud resources (or CloudNets). This solution emphasizes flexibility as well as white and black listing of properties when describing resources. While FLeRD focuses on the formulation of virtual-to-physical resources mapping, no support for modeling network topologies and roles is provided. Integrating NDL and CDL, INDL [12] (the Infrastructure

and Network Description Language) aims at providing technology-independent descriptions of computing infrastructures. Based on ontologies and Semantic Web approaches, INDL’s main goal is to decouple connectivity, functionality, and virtualization of resources so that flexibility is ensured and new types of resources can be added without affecting the existing schema. This solution supports network services and resources description, whereas virtual-to-physical mapping is not considered. A virtual network management information model is

proposed in [5] for managing virtual networks in data centers. In this model, virtual nodes that are instantiated on the same physical node are defined in a group of virtual nodes and the mapping to physical ones is expressed as well. Finally, in [6], Houidi et al. proposed a schema for automated virtual networks provisioning whose objective is to define the

properties of virtual resources and their relationships. In this schema, a network element is considered as the basic building component having functional and non-functional attributes. Even though it is intended to define virtual resources, this schema does not cover all the aspects addressed by our model, such as describing a virtual network as a whole, virtual-to-physical mapping, network services description, and modeling of the relationships between roles and resources/services.

V. CONCLUSIONS AND FUTURE WORK

Virtual networks are federated networks created by aggregating virtual resources that span across multiple network domains. In order to instantiate a virtual network, a virtual infrastructure provider needs to aggregate virtual resources offered by one or multiple physical infrastructure providers, by selecting the resources that meet his/her requirements. To maximize the selection likelihood, the offered resources need to be described in fine-grained manner. In this paper, we proposed an integrated and hierarchical information model, enabling the representation of virtual resources, virtual network services, and virtual network roles, as well as their relationships and hierarchy. Furthermore, we demonstrated the use of this information model using a concrete secure content distribution scenario. As future work, we plan to detail our dynamic resource discovery architecture which relies on the defined model, and build a proof-of-concept prototype to validate its operation.

REFERENCES

[1] N. Chowdhury and R. Boutaba, “Network Virtualization: State of the Art and Research Challenges,” IEEE Communications Magazine, vol. 47, no. 7, pp. 20-26, July 2009.

[2] E. Rosen and Y. Rekhter, "BGP/MPLS IP Virtual Private Networks (VPNs)," RFC 4364, Internet Engineering Task Force, February 2006.

[3] M. El Barachi, N. Kara, and R. Dssouli, “Towards a Service-Oriented Network Virtualization Architecture,” in Proceedings of the 3rd ITU-T Kaleidoscope Event 2010 (K-2010), December 2010, pp. 1-7.

[4] M. El Barachi, N. Kara, and R. Dssouli, “Open Virtual Playground: Initial Architecture and Results,” in Proceedings of the 9th IEEE Consumer Communications and Networking Conference 2012 (CCNC 2012), January 2012, pp. 576-581.

[5] "Network Working Group," IETF, 16 July 2012. [Online]. Available: http://tools.ietf.org/html/draft-okita-ops-vnetmodel-07.

[6] I. Houidi, W. Louati, D. Zeghlache, S. Baucke, "Virtual Resource Description and Clustering for Virtual Network Discovery," in IEEE International Conference on Communications Workshops 2009.

[7] L. Richardson and S. Ruby, “RESTful Web Services”, O’Reilly & Associates, ISBN 10: 0-596-52926-0, May 2007.

[8] RDF Working Group, "Resource Description Framework (RDF).," [Online]. Available: http://www.w3.org/RDF/. [Accessed 07 10 2012].

[9] V. Ham et al. "A distributed topology information system for optical networks based on the semantic web," Optical Switching and Networking, vol. 5, no. 2-3, pp. 85-93, June 2008.

[10] Y. Xu et al. "A Reference Model for Virtual Resource Description and Discovery in Virtual Networks," in Computational Science and Its Applications – ICCSA 2012, vol. 7335, 2012, pp. 297-310.

[11] G. Schaffrath et al., “A Resource Description Language with Vagueness Support for Multi-Provider Cloud Networks," in the 21st International Conference on Computer Communications and Networks (ICCCN 2012).

[12] M. Ghijsen et al., "Towards an Infrastructure Description Language for Modeling Computing Infrastructures," the 10th International Symposium on Parallel and Distributed Processing with Applications (ISPA 2012).

4782