P2P solves the bottleneck of data server transmission in IPTV

Welcome to the IT technology community forum and interact with 2 million of technical staff. With the rapid development of embedded technology over the past few years, streaming media technology, which has a strong market demand, is displayed on the personal computer platform, at present, there is a trend of moving to the embedded platform. Media consumption demands for personal portability and household appliances provide a market opportunity for this platform transfer. I

Welcome to the IT technology community forum and interact with 2 million technical staff> with the rapid development of embedded technology over the past few years, streaming media technology, which shows a strong market demand on the personal computer platform, at present, there is a trend of moving to the embedded platform. Media consumption demands for personal portability and household appliances provide a market opportunity for this platform transfer. I

Welcome to the IT Community Forum and interact with 2 million technicians>

With the rapid development of embedded technology over the past few years, the streaming media technology, which shows strong market demand on the personal computer platform, has gradually shifted to the embedded platform. Media consumption demands for personal portability and household appliances provide a market opportunity for this platform transfer.

IPTV is at the forefront of this consumption revolution. Most people in the industry think that IPTV contains huge commercial benefits, but few people are still involved, one of the reasons is the speed of the network. This article introduces the peer-to-peer (P2P) technology, which is applied to IPTV Network Transmission and can make full use of the upstream and downstream channels of each terminal, vrouters and other hardware are not required to support multicast, which can provide a solution to the IPTV Network Transmission bottleneck problem.

I. Peer-to-Peer Network Technology

1. Peer-to-peer network Introduction

Peer-to-peer network (P2P) is a new network technology that has emerged in recent years. It breaks the conventional network server/client (C/S) pattern in transmission mode and establishes a direct communication mechanism between the client and the client, this communication method also provides ideas for the future development of networks. In a peer-to-peer network, each node serves both as a client and a server of another person. In a sense, each node is in the same status. A peer-to-peer network is a successful expansion of the distributed concept. It distributes the server load in the traditional mode to each node in the network, and each node will undertake limited storage and computing tasks, the more nodes you add to the network, the more resources the node contributes, and the higher the service quality.

On the other hand, the C/S service model still dominates the network. Although the network bandwidth of end users has increased year by year in recent years, the user experience has not improved significantly, this is mainly because the servers are overwhelmed. When the number of users increases sharply, the servers may even be in the danger of crash. At this time, idle link bandwidth is being wasted. P2P network services can fully explore idle network resources and serve more users without additional investment. Therefore, we can see that the advantages of the peer-to-peer network technology are self-evident. It will create great value in future network applications.

2. Basic Structure of the peer-to-peer network

Peer-to-peer networks are generally divided into three types based on their organizational structures.

(1) centralized peer-to-peer network

A centralized peer-to-peer network is not a fully equivalent network. It is an extension of the traditional network mode based on the P2P concept between the C/S mode and the peer-to-peer network. The centralized peer-to-peer network is based on the central directory server. It provides directory query services for various programs in the network, and the transmitted content does not need to go through the central server. The structure of the network is relatively simple, and the burden on the central server is greatly reduced. However, due to the existence of central nodes, it is easy to form a transmission bottleneck and poor scalability, which is not suitable for large networks. However, due to centralized directory management, small network management and control is a feasible solution.

(2) unstructured Distributed Network

The most significant difference between a non-structured distributed network and a centralized network is that it does not have a central server. All nodes access the entire network through communication with adjacent nodes. In a non-structured network, nodes use a packet query mechanism to search for required resources. Specifically, a node sends a query package containing the query content to the adjacent node. The query package spreads across the network in a diffusion manner, if this method is not controlled, messages may flood. Therefore, an appropriate TTL is usually set to decrease in the query process, if the TTL value is 0, no sending will continue.

In this non-structured mode, the organization mode is loose, and the addition and departure of nodes are relatively free. It is easy to find popular content when querying, but if the content to be requested is relatively unpopular, the smaller TTL is not easy to find, and the larger TTL value is easy to cause large query traffic, especially when the network range is extended to a certain scale, even if the limited TTL value is small, the traffic will increase dramatically. However, when there are some so-called class server nodes with rich resources in the network, the query efficiency can be significantly improved.

(3) Structured Distributed Network

Structured Distributed Network is a research achievement Based on Distributed Hash Table (Distributed Hash Table) in recent years. The basic idea is to organize all the resources in the network into a huge table. The table contains the resource keywords and the address of the stored node, split the table and store it in each node of the network. When you search for resources on the network, you can find the nodes where the hash table content corresponding to the keywords is stored, the node contains the Node Address of the required resources, and then the node that initiates the search connects to the corresponding node and transmits the resource according to the address information. This is a technically advanced peer-to-peer network. It is highly structured, highly scalable, and freely joins and leaves nodes. This method is suitable for large networks.

II. Application of peer-to-peer network technology in IPTV

1. IPTV status quo

IPTV and P2P in China have been developing in two different directions. Because IPTV is similar to traditional TV watching modes, it is basically regarded as an upgrade of cable TV (CATV) or digital TV (DVB, therefore, the same central server model is used in the design concept. In the traditional method, the network is basically a one-way channel of broadcasting, that is, the client is unable to send upstream information or has little capacity to send upstream information. In this way, all the service content is sent by the central server, and all the processing work is completely completed by the central server, which cannot implement P2P connections.

To solve the server bottleneck problem in IPTV in the C/S mode, most of the current solutions are to adopt Multicast technology and work with the Content Delivery Policy (CDS) of a central server-Edge server).

Multicast technology is also very popular in recent years. It is superior to the one-to-one multi-point transmission service technology in broadcast networks. In traditional multi-point communication, the server must establish independent connections to each client that requests to connect. Generally, the maximum number of connections that a server can provide is quite limited, therefore, this method is not suitable for media services. In broadcast mode, you do not need to create a corresponding

[1] [2] [3] [4] [5]