Configuring the PC Network

Every computer needs an operating system, including a computer in a network. Due to specific network requires OS network is implemented as a complement to your OS or as a special system based on the server, with the purpose of sharing network resources and use of services. Services such as sending e-mail, messaging, interactive voice or written communication, access local or external Web server, file exchange, or some specific services that network users need, will enable local server in the network or external servers to be used to define network or users in a network must use their resources.

Peculiarities server network operating subsystems are multitasking, background data writing while reading a disc, the duplication of records of the FAT and directories regarding improved reliability, process control transactions (operations on entries in the database on the server drives), email (exchange messages and files), resource sharing networks, prevent conflicts when data access and efficiency.

In order to protect the data entered on the server used double and tandem paired disks (RAID). Dual discs operate on the principle of mirror disk and copy the contents and if there is an error in reading the data from the primary records (e.g. due to damaged disc) OS network is reflected in the copied content and serves the correct information. Powerful OS besides principles mirror reflection using paired duplicate disks, i.e. the files are recorded as a double track on two separate disk. If one disk and file access is busy processing other with identical data. Such an approach is more complex and more expensive, but increases the security and speed of access to data. Of course, every transaction with the data must be monitored simultaneously on both disks are worrying about the network OS. OS will not allow the transaction if an error occurs during a transaction and to preserve the contents of the file before the transaction. That efficiency of the transaction relating to the protection of data is largely determined by the quality of the network OS.

It covers the basic principles of safety data integrity. RAID (Redundant Array of Independent Disks) technology described in the previous Chapters, all present in the design of network servers and newer operating systems such as Windows XP or Windows 7, support the elements of a good deal of network communications, security, and support for RAID system. Although the Windows operating systems on the interface at first glance similar to the workstations and servers, the large differences in the operating systems intended for workstation users (PCs) in relation to the same family of operating systems for servers, client-server concepts. The big difference is in the design of hardware for personal computers and servers. The server must in principle work for years, without interruption, and the design of its hardware device is taken a lot more attention. The difference is already noticeable during the initial contact with the devices, carrying packaged equipment - servers are many times larger and heavier.

Connection between the workstation and the server enables network communications subsystem that manages the hardware resources of the network (e.g. memory) and data flow through the network protocols by default. It also takes care of receiving and transmitting data out of the network. Thus, it allows the integration of different platforms either through modification of data or the use of software tools that are kept on the server, but belong to one type of OS and also use common-shared (SHARE) network resources.

Expanding the network and connect diverse networks to outside servers for this purpose specifically designed as:

Repeater essentially continues the same branch of topology, works as a signal amplifier and recovering on the L2 of OSI model. Bridge connecting systems at L3 and associated networks, the same or different topology. Router allows you to transfer data between networks and acts on the L4 level of OSI model that adapts data for transfer from one system to another.

So, part of a network OS is 'router' aid, whether external (Figure 3.6.10-A) when focusing on the outer 'router' performs and supervises these tasks, or internal (Figure 3.6.10-B) when the server across differently configured network cards and software to connect a variety of network topologies.

Figure 3.6.10 Diagram of a different subnet for the server.

The router can be set and behind the server as the output device in a local area network. In essence, the task router is based on analysis of overdue information (destination address) select the most appropriate route, which will pass the data. Usually in their algorithm is able to work if he determines that it can not establish a common communication path, for example due to a malfunction, try to communicate through others to avoid. In addition to the above can be configured to restrict certain types of unwanted traffic.

Well harmonization of certain types of protocol is of great importance for the stability of the various subnets bound. The local network is often useful to give the possibility to computers and additional devices can exchange data with multiple types of protocols. For example, the Internet will use TCP / IP, and to retrieve the print server in the local network will be used NetBEUI or IPX / SPX, which are somewhat easier and quicker in the local paper.

Software for connecting stations establishes coherence (correlation) between the command station and OS server OS commands to the command server OS becomes interpreted in the same way. OS becomes in this context is completed. Therefore, the user can directly access the network using network commands and user programs that are adapted type OS station. Accordingly, the network OS has support for different platforms to different work stations with their own OS could connect to the network.

Each network must have at least one server. They can be for the same purpose or for a variety of purposes integrate more into the network and 'bind' their mutual work. Network concepts based on multiple servers (for example one works and the other he is a true copy) of fundamentally changing the stability of the network.

Figure 3.6.11 The concept of network with one or more servers.

If a network system based on the work of just one server, which is also the holder of the network, its incorrectness cease to function all users connected to it. Network and server are on the same level. The role and position of supervisor-administrator of the network and the users are not shown the same solutions.

If a network system based on multiple servers, the network represents a higher level in the hierarchy of a set of servers. The network is based on the number of 'pillars' and shooting one will cause the destruction of the whole system, which is an obvious contribution to the stability of the system. User-level is above the level of the server, which makes it completely independent of the state of individual elements of the system, namely the 'failure' of one server does not affect the user. User content at the network level and not at the server level - NDS System (Network Directory Service). NDS is a global database that contains information about the entire network system. Of course, the file server will create and print server that will take care of the printing resources. A system for monitoring transactions - TTS (Transaction Tracking System), will allow to realize some unfinished process as entry or transfer into a database terminated due to failure workstations, and thus ensures continuous data integrity, and correctness.