Computer network

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Typical library network, in a branching tree map and controlled access to resources

A computer network is a group of computers connected to each other electronically. This means that the computers can "talk" to each other and that every computer in the network can send information to the others. Usually, this means that the speed of the connection is fast - faster than a normal connection to the Internet. Some basic types of computer networks include:

  • A local area network (often called a LAN) connects two or more computers, and may be called a corporate network in an office or business setting.
  • An "internetwork", sometimes called a Wide Area Network (because of the wide distance between networks) connects two or more smaller networks together. The largest internetwork is called the Internet.

Computers can be part of several different networks. Networks can also be parts of bigger networks. The local area network in a small business is usually connected to the corporate network of the larger company. Any connected machine at any level of the organization may be able to access the Internet, for example to demonstrate computers in the store, display its catalogue through a web server, or convert received orders into shipping instructions.

Microsoft Windows, Linux and most other operating systems use TCP/IP for networking. Apple Macintosh computers used Appletalk in the past, but it uses TCP/IP now.

To set up a network an appropriate media is required. This can be wired or wireless. Twisted-pair, co-axial or fiber-optic are examples of cable and infra-red, blue-tooth, radio-wave, micro-wave etc. are wireless media used for networking. When you are working with a mere LAN, computers, media and peripherals are sufficient. But when you are working with a wider range you have use some additional devices like bridge, gateway or router to connect different small or large networks. And obviously a protocol must be maintained.

To set up a network you have to select an appropriate topology to arrange the hardware devices using the media. Topologies generally used are bus-topology, ring-topology, star-topology, tree-topology, object-oriented topology etc. Among these star-topology and tree-topology are most popular nowadays.[source?]

Network models[change | edit source]

Example of communication in network model

The network communication technology as one complete would be very difficult to implement. To simplify it, we must implement various layered network models such as ISO/OSI. All network models are based on the same principle. The upper layer uses services of the lower one and provides services for the upper one. The specific layer can only communicate with the same layer on the destination device.

OSI model[change | edit source]

OSI (Open Systems Interconnection) is a 7-layer network model specified by ISO (International Organization for Standardization) norm and is widely used all around the world. The concept of a seven-layer model was provided by the work of Charles Bachman[1], Honeywell information Services. Various aspects of OSI design evolved from experiences with the ARPANET, NPLNET, EIN and CYCLADES networks and the work in IFIP WG6.1.

Data unit Layer Function
Data Application Network process to application
Presentation Encryption, decryption and data converting
Session Managing sessions between applications
Segments Transport End-to-end connection and reliability
Packets (datagrams) Network Path determination and logical addressing
Frame Data link Physical addressing
Bit Physical Signal and binnary transmission

Layer 1[change | edit source]

Digital modulation:
16-QAM with example constellation points.
Analog modulation:
FM - frequency
AM - amplitude

The physical layer defines electrical and physical specifications for devices. It also specifies modulated and baseband transmission.

Baseband[change | edit source]

Baseband is transmission of digital data in their raw form (1001 1101 1010 0011). This allows very fast and reliable communication over short distances; however, due to the realistic electrical properties of used media, the range of baseband transmission is very limited and decreases with increasing speed. Baseband technology is frequently used on LAN.

  • UTP cable – max 100 m on 100 Mbit/s speed without repeater
  • Optical fiber – max 1 km on 100 Mbit/s speed without repeater

Typical technology: Ethernet

Modulated transmission[change | edit source]

In telecommunications, modulation is the process of conveying a message signal, for example a digital bit stream or an analog audio signal, inside another signal that can be physically transmitted. The device that provides modulation of the baseband signal is called a modulator, and the device that provides demodulation of the modulated signal back to baseband is called a demodulator. Today, the modulator and demodulator are integrated into one device called Modem (modulator-demodulator). Frequently used on WAN, WLAN, WWAN.
Typical technology: WI-FI, ADSL, cable TV connection (CATV)

Layer 2[change | edit source]

The data link layer provides the functional and procedural means to transfer data between network entities and to detect and possibly correct errors that may occur in the physical layer.

Layer 3[change | edit source]

The network layer provides the functional and procedural means of transferring variable length data sequences from a source host on one network to a destination host on a different network using the IP address.

IP address[change | edit source]

An Internet Protocol address (IP address) is a numerical label assigned to each device (e.g., computer, printer) participating in a computer network that uses the Internet Protocol for communication. At present there are two versions of protocols in use – IPv4 and IPv6.

  • IPv4 uses 32-bit addressing which limits address space up to 4294967296 (232) possible unique addresses.
Example: IP-192.168.0.1 mask-255.255.255.0 means that the network address is 192.168.0.0 and the device address is 192.168.0.1
  • IPv6 uses 128-bit addressing which limits address space up to 2128 possible addresses. It is deemed sufficient for the foreseeable future. Full IPv6 support is still in the implementation phase.

Layer 4[change | edit source]

The transport layer provides transparent transfer of data between end users, providing reliable data transfer services to the upper layers. The Transmission Control Protocol (TCP) and the User Datagram Protocol (UDP) of the Internet Protocol Suite are commonly categorized as layer-4 protocols within OSI.

  • TCP (transmission control protocol) provides reliable, ordered delivery of a stream of bytes from a program on one computer to another program on another computer. TCP is used for applications that strictly require reliable transfer (e-mail, WWW, file transfer (FTP), ...)
  • UDP (user datagram protocol) uses a simple transmission model without implicit handshaking dialogues for providing reliability, ordering, or data integrity. UDP is used in applications where we require reduced latency over reliability (stream videos, VOIP, online games,...)

Layers 5-7[change | edit source]

Commonly united into one layer in simplified network models, its main purpose is to interact with applications, encrypting and establishing dedicated connections if necessary.

Networking terms[change | edit source]

Latency[change | edit source]

Latency, incorrectly called ping, is a value that measures how much time packets need to travel to their destination. It is measured in miliseconds (ms). The tool that measures latency is called ping, commonly using special ICMP packets which are smaller than standard data packets so they do not weight the network by their presence.

  • Immediate latency is measured every X seconds and immediately displayed. Its value constantly changes due to the natural properties of packet-switching network technology. High latency peaks have negative effects on most network applications which can adapt to average latency by allocating corresponding size of memory as buffer. High latency peaks lead to emptying of this buffer and to temporary freezing of applications. This freezing is commonly called lag.
  • Average latency is the sum of immediate latency measured Y times every X seconds divided by Y. Average latency is used to estimate size of the buffer, mainly because it does not change so often. Buffer enables some applications such as stream videos to run smoothly even with high average latency, but it cannot protect us from high latency peaks.
ADSL frequency plan.
Upstream + downstream = network bandwidth

Capacity (bandwidth)[change | edit source]

Capacity is a measure of transfer capacity of a network and is measured in bits per second (bps or b/s), today commonly Mbps or Mb/s. It shows us how many data units are transferred each second. At present, the average bandwidth is far higher than is necessary and it is not a limiting factor in most cases.

  • Uplink is how much bandwidth is used for transferring data from user to server (usually lower for end users).
  • Downlink is how much bandwidth is used for transferring data from server to user (usually higher for end users).

Broadcast[change | edit source]

Broadcast is a special transmission that is not addressed to a single device but to all devices in specific network. It is mostly used to automatically issue IP addresses to devices by a DHCP server and to create an ARP table that maps the network and speeds up the traffic.

References[change | edit source]

  1. [1] at www.computerhistory.org.