Authors: Jiann-Liang Chenz, Szu-Lin Wuy,Yang-Fang Li, Pei-Jia Yang,Yanuarius Teofilus Larosa 2011 7th International Wireless Communications and Mobile.

IMS Cloud Computing Architecture for High-Quality Multimedia Applications

Authors:Jiann-Liang Chenz, Szu-Lin Wuy,Yang-Fang Li, Pei-Jia Yang,Yanuarius Teofilus Larosa

2011 7th International Wireless Communications and Mobile Computing Conference (IWCMC), On page(s): 1463 - 1468, July. 2011 報告 : 陳昶安 MA1G0107

Outline

Introduction

Background Knowledge

Proposed IMS cloud QOS mechanism

System design and performance analysis

Conclusion

2

Introduction

The IP Multimedia Subsystem (IMS) supports heterogeneous networking and provides Quality of Service (QoS) policy.

For mobile computing, an open platform is required to access multimedia services.

All of the above issues must be addressed when developing next-generation wireless communications technologies and applications.

3

Introduction

Figure 1 illustrates a IMS cloud architecture in a heterogeneous network.

4

Introduction

The IMS architecture which is based on the 3GPP (3rd Generation Partnership Project) system, defined QoS policy module, which consists of Policy and Charging Rules Function (PCRF), Policy and Charging Enforcement Function (PCEF).

The cloud computing architecture is divided into three layers, Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS).

5


Heterogeneous Networks

• The heterogeneous network is a next-generation communication network that provides wireless links via 3G, WLAN or WiMAX in overlapping networks.

• A 3G supports multimedia communication and is expected to evolve into All-IP technologies.

• The WiMAX wireless broadband technology provides long-distance last-mile access to high bandwidth.

6


The IEEE 802.11x protocols for wireless local area network (WLAN) use the same basic protocol for over-the-air modulation techniques.

In heterogeneous networks, 3G can support a high-mobility environment.

The WiMAX supports mobile terminals moving at speeds up to 120Km/h.

7


Cloud Computing

• The cloud computing architecture is divided into three layers.

• The bottom layer is Infrastructure as a Service (IaaS), which has a service-oriented architecture.

• The middle layer is Platform as a Service (PaaS), a service platform that developer can use to deploy their own applications.

8


The top layer of the Software as a Service (SaaS) provides services so that each user can access services according to their requirements.

Hadoop is a software platform for a java-based development environment that has high scalability, high efficiency, and high reliability.

The two components of Hadoop are Hadoop Distributed File System (HDFS) and MapReduce.

9


IP Multimedia Subsystem (IMS)

• The purpose of IMS is to provide services in a 3G network infrastructure and to enable QoS exchange in different networks.

• The 3GPP-defined IMS core component is Call Session Control Function (CSCF).

• The components are divided into P-CSCF (Proxy-CSCF), I-CSCF (Interrogating-CSCF) and SCSCF modules (Serving-CSCF).

10


The P-CSCF is the first contact point in IMS.

The P-CSCF is responsible for authentication requests, which is forwarded to the specific target.

The ICSCF is the contact point between network operators and UEs (User Equipment).

The S-CSCF is the core component of IMS, which is located in attributed network for UE session control and registration services.

11


Figure 2 illustrates the proposed IMS Cloud QoS system architecture.

12


The lower layer model, IaaS is the development platform used for building the IMS system.

The PaaS provides a platform for system administrators to develop the Hadoop framework needed for distributed file system and MapReduce technology.

The upper layer model is SaaS, which enables users to obtain information and services via cloud computing systems.

13


Figure 3 illustrates the proposed system module, which includes IaaS, PaaS and SaaS.

14


Infrastructure as a Service (IaaS)

• In IaaS, a service is accessed through different networks in an IMS environment.

• The IMS is implemented through an open and standard architecture to provide multimedia applications.

• The IMS framework includes the Policy and Charging Rules Function (PCRF) and Policy and Charging EnForcement point (PCEF) components needed for QoS policy implementation.

15


Figure 4 illustrates a complete IMS QoS framework architecture, in which the policy repository by the open source of XDMS server is used to store definitions of QoS policy files.

16


Platform as a Service (PaaS)

• The high fault tolerance of HDFS enables it to deploy on low-cost hardware devices.

• The HDFS is suitable for massive data and provides high throughput.

• MapReduce is a parallel programming model for data processing and is used by Google for large-scale data processing in distributed computing environments.

17


Software as a Service (SaaS)

• This layer is implemented by using the Hadoop SaaS system for extracting the user wants to access information and applications to achieve the service function.

• For instance, users can access applications through Web Services such as multimedia (video streaming), management (user process management) and social network applications.

18


Figure 5 shows the cloud computing service architecture.

19


Figure 6 illustrates the test scenario, which included 3G, WiMAX and WiFi access technologies.

20


System Implementation1) Hadoop system design and operation:

• One mainly machine, called hdp-1, is the Master, and another two machines, hdp-2 and hdp-3, are the Slaves.

• This study utilizes the Representational State Transfer(REST) Web Services to get Hadoop services.

21


• The user interface is shown in Figure 7.

22


Figure 8 shows the installed system.

23


2) IMS core design and operation:

• Figure 9 illustrates the established IMS environment.

24


This study sets up IMS components in the master machine (hdp-1).

Clients then communicate by using SIP protocol to access the IMS network.

The system defines all domain policies rule.

25


The four CSCFs modules are started as shown in Fig. 10.

26


Figure 11 shows the Policy Control Management.

27


Figure 12 shows how an Android phone is used to access the cloud computing portal site by QR code and how services can be accessed through the cloud computing interface.

28


Performance Analysis • Figure 13 compares the results with and without cloud computing system

when a client request is set to retrieve 128MB.

29


Figure 14 shows the system throughput for VoIP, video streaming and Web browser.

30

Conclusion

• The proposed IMS Cloud QoS mechanism has the following three layers:

Infrastructure as a Service (IaaS) layerPlatform as a Service (PaaS) layerSoftware as a Service (SaaS) layer

• The IaaS layer mainly accesses heterogeneous networks such as UMTS, WLAN and WiMAX through the IP Multimedia Subsystem.

31

Conclusion

The PaaS layer manages the cloud computing system in the Hadoop Distributed File System and the MapReduce mechanism for analyzing user preferences.

The SaaS layer allows users to access desired applications such as the Web, social networks and management services.

The system implementation shows that throughput capacity with cloud computing is higher than without cloud computing when numerous users request services.

32

Authors: Jiann-Liang Chenz, Szu-Lin Wuy,Yang-Fang Li, Pei-Jia Yang,Yanuarius Teofilus Larosa 2011 7th International Wireless Communications and Mobile.

Documents