Dynamic resource management has become an active area of research in the Cloud Computing paradigm. Cost of resources varies significantly depending on configuration for using them. Hence efficient management of resources is of prime interest to both Cloud Providers and Cloud Users. In this work we suggest a probabilistic resource provisioning approach that can be exploited as the input of a dynamic resource management scheme. Using a Video on Demand use case to justify our claims, we propose an analytical model inspired from standard models developed for epidemiology spreading, to represent sudden and intense workload variations. We show that the resulting model verifies a Large Deviation Principle that statistically characterizes extreme rare events, such as the ones produced by “buzz/flash crowd effects” that may cause workload overflow in the VoD context. This analysis provides valuable insight on expectable abnormal behaviors of systems. We exploit the information obtained using the Large Deviation Principle for the proposed Video on Demand use-case for defining policies (Service Level Agreements). We believe these policies for elastic resource provisioning and usage may be of some interest to all stakeholders in the emerging context of cloud networking.

There is an increasing demand for the technology of content distribution, by which each user can request desired content through a network. Because of the low efficiency of existing systems, we proposed a new block transfer type video distribution system called Burst VoD. The Burst VoD system aggressively utilizes multicasting, and divides the content data into a mass of block files, which it periodically transmits to a terminal through a high-speed network, using a higher rate than the playback speed. However, by using the scheduling algorithm of the Burst VoD system, when users request the same content from different periods, the VoD server repeatedly transmits the same block files in different periods. In this paper, we propose an advanced scheduling algorithm based on the Burst VoD system to improve its multicasting efficiency. In addition, we propose a multi-channel BurstVoD in order to reduce the interface bandwidth of client.

In this paper, a new video broadcast protocol is proposed for video-on-demand (VoD) in shared environment. The new protocol is developed by modifying the first segment delivery scheme for the skyscraper protocol using the idea of patching. The results show that the start-up latency for users is greatly reduced when using our new protocol.

A systematic continuous data placement scheme on distributed multi-zone disks is developed for video on demand. The proposed scheme makes use of constant read time concept, i. e. , each video stream has the same access time in each service round when serving the request. The developed scheme maximizes not only the averaged data transmitted rate, but also the number of simultaneous accesses. The scheme consists of the following components. First we developed an algorithm that reorganizes the multi-zone disk into several logical zones in the sense that the averaged disk throughput is maximum. Second, a sequential data access method was developed that takes disk loading balance into account. Thus, at each service round, the total amount of data transmitted is a constant. Third, we introduce the idle round technique to reduce the buffer size required at the client site for VBR video stream. As a result, admission control is enforced in an efficient manner. Finally we perform experimental tests to evaluate the performance of the proposed scheme. The results indicate the value of the proposed scheme.

Video on demand is regarded as one of the most important applications in future broadband networks. Although several trial runs in various parts of the world do not bring instant success, its inherent requirements for bandwidth and bounded delay make it a suitable candidate for studying various networks and protocols in real-time applications. Recently, the Hybrid Fiber Coaxial (HFC) networks emerged as network architecture for bringing broadband services to the home. The co-axial cable network uses a tree-like architecture and requires complex protocols for synchronizing and sharing uplink communications among many users. This paper proposes that a dual bus network may be a simpler and cost effective alternative. We implement an interactive VOD (IVOD) system as an example to demonstrate its simplicity and suitability as a community network. The system uses client/server architecture with guaranteed bandwidth and bounded delays. Simulations are conducted to test the network load and response time for different number of users. The results indicate its feasibility and the delay time can be further reduced by releasing bandwidth during the pause operation.

In this paper, we study the optimal allocation of multimedia files in distributed network systems. In these systems, the files are shared by users connected with different servers geographically separated, and each file must be stored in at least one of servers. Users can access any files stored in any servers connected with high-speed communication networks. Copies of the files accessed frequently are to be stored in several servers that have databases. So, it is one of the most important problems how to assign the files to servers in view of costs and delays. Considering these problems in heterogeneous network environments, we present a new system model that covers wide range of multimedia network applications like VOD, CALS, and so on. In these systems, it is obvious that there is trading-off relationship between costs and delays. Our objective is to find the optimal file allocation such that the total cost is minimized subject to the total delay. We introduce a 0-1 integer programming formulation for the optimization problem, and find the optimal file allocation by solving these formulae.

In this paper, we present a distributed RAID style video server that addresses the problem of increasing video stream supplying capability in VOD systems. Distibuted RAID Stylemeans extending the RAID method usually applied to disks to the servers, so as to achieve improvements in performance and reliability. The great advantage of this architecture is that the linear performance improvement can be achieved by adding inexpensive servers of a diffusion model. We have implemented the RAID0 style and RAID4 style systems. A number of new features such as the array configuration method or the striping buffer are applied to them. A precise performance evaluation was made in a near practical environment, with more than 40 video streams. We tested the system, in both normal and degraded mode (server failure) and by checking frame drop rates as well as observing picture quality. An evaluation by simulation was also conducted. The results of the above evaluations agreed well with the calculation value, to confirm the feasibility of this architecture.