计算机网络常识（很有用）
General Knowledge
A network consists of several computers linked by communication lines. The machines can function independently; but their activities can also be coordinated. In the supermarket network, the checkout stations collect data, the minicomputers perform store-level processing, and the large mainframe handles chain wide inventory and distribution tasks．In another distributed network, the accounting, sales, production, and engineering departments of a busy-ness firm might have their own computers, with althea machines linked to a centralized mainframe housing the organizations database. With such a system, the individual departments have the flexibility to do their own processing, while the central machine en-sores their of use common data, and helps to monitor and control their activities。
Although the physical separation of the communicating computers may vary considerably from one type of application to another，in general a computer communication network can be described: at the heart of any computer communication networks is the data communication facility which, as has just been outlined ．may be a PSDN，a private LAN or perhaps number of such networks interconnected together. However ,irrespective of the type of data communication facility, an amount of hardware and softwares required within each attached computer to handle the appropriate network-dependent protocols．Typically, these arc concerned with the establishment of communication channel access to the network and with the controls of the flow of messages across this channel. The provision of such facilities is only part of the network requirements, however, since in many applications the communicating computers may be different types ．This means that they may use differ-net programming languages and, more importantly, different forms of data representation interface be-teen user application programs, normally referred to as application processes or Apes，and the underlying communication services may be different. For example, one computer may be a small single user computer, while another may be large multi-user system。
In the earlier days of computer communication, these issues meant that only closed communities of computers (that is, from the same manufacture) could communicate with each other in a meaningful way. IBM's Systems Network Architecture（SNA) adder’s Digital Networks Architecture (DNA) are just two examples of communication software pack-ages produced by manufacturers to allow their sys-teems to be interconnected together. These proprietary packages, however，of which there are still many inexistence, do not address thus problems of universal interconnectivity, or open systems interconnection. Inane attempt to solve this problem, the ISO, in the late1970s，formulated a reference model to provide common basis for the coordination of standards de-elopements and to allow existing and evolving standards activities to be placed into perspective with one another. The ultimate aim was to allow an application process in any computer that supported the applicable standards to freely communicate with an application process in any other computer supporting the same standard, irrespective of its origin of Manu-facture ．This model was termed the ISO Reference Model for Open System Interconnection。
It should be stressed, however, that this model isn't concerned with specific applications of computer communication networks．Rather，it is concerned with the structuring of the communication software that is needed to provide a reliable, data transparent, communication service (which is independent of any specific manufactures equipment or conventions) to support a wide range of applications。
Networks are not limited to such hierarchical structures, however. For example, many large or-generations with computers spread all over the world link their machines t0 form a ring network. Obviously, with all the computers linked, the various divisions and offices can exchange information. Less obvious is the backup provided by such a network; if one computer fails, its work can be switched to the others．
Not all networks are private. For example, any one with a microcomputer, a modem, the right software, and a telephone can link to any number of computers and access a variety of data ranging from stock quotations to library catalogs, to electronic bulletin boards. More sophisticated systems even allow users all over the world to exchange electronic mail ．Such as，we have just begun to tap the potential of computer networks．
翻译：
网络常识
网络是由通讯线路连接起来 的一些计算机组成。这些计算机既有自己独立工作的能力，又能互相协调地工作。在超级市场的网络中，结帐台收集数据，小型 机完成分店和商务公司的处理，而大型的主机则处理网络内的货物清单和分配任务。在另一个分布式网络中，会计、销售、产品 和程技术部门都可以拥有自己的计算机。所有这些计算机和一台中央机相联，该主机的机房内还有数据库。对于这样一个系统，每个独众部门部有处理自己工作的灵活性，而中央主机则保证各个独立部门使用的公共数据，并协助监督和控制它们的活动。
虽然相互通信的计算机间的实际距离因应用类型不同有很大变化，但一般来讲，可描述如下：任何一个计算机通信网络的核心是数据通信设备，这些设备 ；可以是PSDN，专用LAN或很多这种网络的互联系统。但是如果不考虑数据通信设备的类型，那么为处理相应的与网络有关的协议，需在联网的计算机上配备很多硬件和软件。一般来讲这些都与跨网通信信道的建立，与通过该信道的信息流控制有关。提供这种设备仅是网络要求的一部分，但是在很多应用中，通信的计算机类型可以不同。这表明它们可以使用不同的编程语言，而更重要的是可以使用不同的数据表示形式。同时，由于各计算机使用不同的操作系统，因而用户应用程序（通常叫做应用处理程程序或AP）与基本通信服务程序之间的接口也不同。例如，一台计算机可能是中小型单用户机，而另一台可能是大型多用户系统。
在计算机通信的初期，只是计算机封闭系统（即同一厂家的计算机）间相互通信。典型的互联系统通信软件包是IBM的系统网络结构SNA和DEC的数字网 络结构DNA。这些专用软件包尽 管现在仍用得很多，但不具有通用互联性，即不具有开放系统互联性。为了解决这一问题，国际标准化组织ISO在70年代后期提 出了一个参考模型，该模型为协调标准的开发提供了一个公共依据，并且使得现有的和发展中的标准工作相互协调。其最终目标是使支持这些可用标准的任何计 算机能与支持相同标准的其它计算机在一个应用处理程序上随意通信，而不管使用何种计算机。这一模型称作为ISO开放系统互联参考模型。
这一模型与计算机通信网络的具体应用无关。它只是给出用于可靠的、数据透明的通信服务 的通信软件结构，支持广泛的应用而与任何厂家的设备和规定无关。
然而，网络不仅只限于这种层次结构。例如，拥有一些计算 机的许多大型组织机构遍布于全世界，他们把这些计算机相互连 接起来，形成一个封闭的网络。 显然，各级部门和办公室都能使 用已经连接起来的计算机交换信 息。不那么明显的是，这种网络 提供的备份功能，如果一台计算 机发生故障，它的工作能转换到 其它的计算机上去完成。
网络并不是专用的。例如， 任何一个人只要拥有一台微机、一台调制解调器、再加上适当的软件和一部电话，他就能与任意数量的计算机相连，其范围可以 从股票行情到图书馆目录，知道电子布告板。如此等等，我们也只是刚开始发掘计算机网络的潜力。
