QoS-aware Virtual Machine Scheduling for Video Streaming Services in Multi-Cloud

QoS-aware Virtual Machine Scheduling for Video Streaming Services in Multi-Cloud

Abstract:

Video streaming services are trending to be deployed on cloud. Cloud computing offers better stability and lower price than traditional IT facilities. Huge storage capacity is essential for video streaming service. More and more cloud providers appear so there are increasing cloud platforms to choose. A better choice is to use more than one data center, which is called multi-cloud. In this Project a closed-loop approach is proposed for optimizing QoS and cost. Modules of monitoring and controlling data centers are required as well as the application feedback such as video streaming services. An algorithm is proposed to help choose cloud providers and data centers in a multi-cloud environment as a video service manager. Performance evaluation of the algorithm is included with different video service workload. Compared with using only one cloud provider, dynamically deploying services in multi-cloud is better in aspects of both cost and QoS. If cloud service costs are different among data centers, the algorithm will help to make choices to lower the cost and keep a high QoS

Introduction

Cloud computing is changing more and more services on Internet. In the area of IaaS, Amazon is the most popular cloud provider, but more and more providers are coming into this area. The numbers of cloud providers will increase explosively in future. Netflix is a video streaming service provider and based on Amazon EC2. It has been proved that a video service based on cloud computing is feasible. But with more cloud providers, how to choose from the providers is becoming increasingly important. Different cloud providers may charge a different price and  support different service item. One cloud provider may have several data centers to choose. The position of

data center is also important for IO type service like streaming video. The quality of service (QoS) will decrease if the data center is far from the end users. In such a multi-cloud environment, applications based on cloud should make choices of how to use these resources. Security in cloud computing is also very important. Lots of works[3,4] have been done to resolve this problem. In multi-cloud, security problem is more important and difficult. With such standard security management, cooperation in multi cloud providers are realizable

Existing Problem Statement

In cloud computing, economics are becoming critical important for both the cloud providers and users Video streaming technique has been developed for several years and can resolve lots of problems for the online video demand. But on a large scale situation, more targeted development and optimization are required. authors introduced key issues on video streaming. Application-layer QoS are specially discussed because it is very important in video application. CDN (content delivery network) is also a very important way to lifting the quality of video service. It is a buffer-like service which can support content delivery need. By CDN only, lots of problems are not solved very well, so some related technique based on CDN are develop.

the QoS for voice and video streaming on Internet. The QoS is affected by the transition delay and packet lost rate

Proposed System

we mainly concern the situation of multi cloud. There are several data center in several places and in each data center, we can use an elastic computing resources. In authors opinion, cloud computing is the trend of the network. More and more small and medium enterprises will choose cloud computing to build their network services instead of buying lots of facilities and employing lots of IT staff to managing them. But with single cloud provider, the network reliability and the price will be a potential risk. A mature large-scale service cannot build their service on one cloud provider.

The main point of this Project, is how to improve the  quality of service and lower the cost in the multi cloud environment  In cloud computing, VM (virtual machine) is the unit of service provided for the users. When the service need more computing ability, users can ask for more VM. In one data center, the network bandwidth is wide enough so the data transaction between VMs are very fast and cheap. For the video service, the system will store a copy of video data in each data center and all the VMs in this data center will share this copy to provide service.

The internet out of data center is more complex. When the user is far from the server in data center, the quality of video service will decrease, because the delay time and packet loss rate will increase. At the same time, because of the retransmission and artificial refresh operation, the press on the service will also increase.

So if there are lots of users around somewhere, a new data center nearby will help resolve the problem. But transferring the data to the new data center and renting

storage space will cost a lot.

 

Conclusions

In this projects describe an algorithm of configure  resources for a video stream service in the multi-cloud environment. Cloud providers are becoming more and more along with the technique developing. For a mature large-scale service, choosing more than one data center is a good choice. This algorithm is used to configure storage and VM resources in this situation. The main contribution of authors includes 2 points. First, authors described the algorithm and realized it. Second, authors made a simulation to validate the effectiveness of this algorithm.

Requirement Analysis:

System Requirements:-

Ø  Language                          :  Java1.5

Ø  Front End                          :  Java Swing

Ø  Back End                          :  SQL Server 2000

Ø  Operating System             :  Windows XP.

Hardware Requirements:-

Ø  Hard disk                          : 60GB

Ø  RAM                                 : 1GB

Ø  Processor                           : P IV