Network card

The network card, also known as the network card, network adapter, LAN adapter, Physical network interface, or its terms in English network interface card or network interface controller (NIC), whose literal translation from English is “network interface card” (TIR), is a hardware component that connects a computer to a network computing and that makes it possible to share resources (such as files, entire hard drives, printers, and the Internet) between two or more computers, that is, in a computer network.

Implementation

ATM network interface.

10 Mbps ISA Network Card.

ISA network card, 400 Mbps, with RJ-45 connectors for 10BaseT, AUI and 10Base2.

Label of a UMTS router with MAC addresses for LAN and WLAN modules

PCI network card of 10 Mbps.

An IPv4 address both in point-decimal notation and in binary code

Decomposition of an IPv6 address of hexadecimal representation to its binary value.

Early network interface cards were commonly implemented in expansion cards that plugged into a computer bus. The low cost and ubiquity of the Ethernet standard made it possible for most modern computers to have a network interface built into the motherboard. Newer server motherboards may even have dual network interfaces built in.

Ethernet capabilities are now built into the motherboard chipset or implemented via a low-cost, dedicated Ethernet chip, connected via the PCI bus (or the newer PCI Express), so no Ethernet connection is required. network card separately unless additional interfaces are needed or another type of network is used.

Modern Network cards offer advanced features such as interfacing interrupt and DMA to host processors, support for multiple receive and transmit queues, partitioning across multiple logical interfaces, and processing of network traffic in driver, such as the TCP offload engine.

The NIC may use one or more of the following techniques to indicate the availability of packets to transfer:

• Pollingwhere the CPU examines the state of the peripheral under the control of the program.
• IRQ-E/S controlledwhere the peripheral alerts the CPU that it is ready to transfer data.

In addition, NICs may use one or more of the following techniques to transfer packet data:

• Input/output programmedwhere the CPU moves the data to or from the NIC to memory.
• DMAwhere any other device other than the CPU assumes the control of the system bus to move data to or from the NIC to memory. This eliminates the CPU load, but requires more logic on the card.

In addition, a packet buffer on the NIC may not be necessary and may reduce latency. There are two types of DMA:

• DMA of third partiesin which a DMA controller other than the NIC makes transfers.
• Bus mastering, where the NIC itself makes transfers.

An Ethernet network card typically has an 8P8C socket where the network cable is connected. Older NICs also provided BNC, or AUI, connections. Some LEDs inform the user if the network is active and if data transmission is taking place or not. Ethernet network cards typically support 10 Mbit/s, 100 Mbit/s, and 1000 Mbit/s Ethernet. Such cards are designated as "10/100/1000", which means that they can support a notional maximum transfer rate of 10, 100 or 1000 Mbit/s. 10 Gbit/s NICs are also available

Purpose

The NIC implements the electronic circuits necessary to communicate over a computer network, whether using cables such as Token Ring, Ethernet, fiber, or wireless such as Wi-Fi, it is therefore a physical layer device and one of The data link layer provides physical access to a network medium and, for IEEE 802 and similar networks, provides a low-level addressing system through the use of MAC addresses that are assigned exclusively to network cards.

This provides a foundation for a complete network protocol stack, allowing communication between small groups of computers on the same local area network (LAN) and large-scale network communications via routable protocols, such as Internet Protocol (IP).

Although other network technologies exist, IEEE 802 networks, including Ethernet variants, have achieved near ubiquity since the mid-1990s.

MAC address and IP address

Each network card has a unique 48-bit hexadecimal identification number assigned by legal hardware manufacturers called the MAC address (Media Access Control; access control to the medium) also known as the physical address which is independent of the network protocol being used. These unique hardware addresses are administered by the IEEE, Institute of Electronic and Electrical Engineers. The first three octets (24 bits) of the MAC number identify the specific vendor and is known as an OUI number (Organizationally Unique Identifier), designated by IEEE, which combined with another 24 bits make up the complete MAC address.

On the other hand, an IP address is a number that identifies, in a logical and hierarchical way, a network interface (communication/connection element) of a device (computer, tablet, laptop, smartphone) using the IP (Internet Protocol) protocol, which corresponds to the network layer of the TCP/IP model. The IPv4 protocol is currently used and the IPv6 protocol is being integrated very slowly.

• The numbering of a MAC address is as follows: "B7:F0:A6:D3:E9:99".
• The numbering of an IPv4 address is as follows: "192.168.1.1".
• The numbering of an IPv6 address is as follows: "0000:0000:0000:0000:00:00:FFFF:C0A8:0101".

Types of network cards

A NIC ISA-16 4/16 Mbit/s for networks Madge Token Ring.

There are no connectors good if there are and those network connectors are: BNC (for coaxial cable); and RJ-45 (for braided pair cable).

Intel 82574L Gigabit Ethernet NIC, a PCI Express ×1, which provides two hardware reception queues.

PCI expansion card for Wi-Fi wireless network.

There are different types of network cards, boards or adapters, depending on the type of cabling or network architecture:

• Token Ring
• ARCNET
• Ethernet
• Wi-Fi

Token Ring

Token Ring network cards are practically obsolete, due to the low speed and high cost compared to Ethernet. They had a DB-9 connector. The RJ-45 connector was also used for NICs and MAUs (Multiple Access Unit), which was the core of a Token Ring network.

Arcnet

ARCNET network cards used mainly BNC connector and/or RJ-45 ports.

Ethernet

Network cards for Ethernet use connectors:

• RJ-45 (Registered jack): 10/100/1000,
• BNC (Bayonet Neill-Concelman10,
• AUI (Attachment Unit Interface10,
• MII (Media Independent Interface): 100,
• GMII (Gigabit Media Independent Interface): 1000

The most common case is that of the card with the RJ-45 connector, although during the transition from the majority use of coaxial cable (10 Mbit/s) to twisted pair cable (100 Mbit/s) cards with BNC and RJ-45 connectors, and even BNC / AUI / RJ-45 (in many of them you can see the unused connectors silkscreened).

With the entry of Gigabit networks and the frequent presence of several computers in homes, cards and motherboards (with integrated NICs) with 2 and up to 4 RJ-45 ports are beginning to be seen, which were previously reserved for servers.

They can vary depending on the transmission speed, usually 10 Mbit/s or 10/100 Mbit/s. 1000 Mbit/s, known as Gigabit Ethernet and in some cases 10 Gigabit Ethernet are also used, also using twisted pair cable, categories: 6, 6a and Cat 7, which operate at higher frequencies.

The speeds specified by the manufacturers are theoretical, for example those of 100 Mbit/s can really reach a maximum of 78.4 Mbit/s.^{[citation required]}

WiFi

NICs are also wireless cards (wireless), which come in different varieties depending on the standard to which they conform, usually 802.11b, 802.11g and 802.11n. The most popular are 802.11b that transmits at 11 Mbit/s (1.375 MB/s) with a theoretical distance of 100 meters and 802.11g that transmits at 54 Mbit/s (6.75 MB/s).

The real transfer speed that a Wi-Fi card with protocol 11.b can reach is about 4 Mbit/s (0.5 MB/s) and those with protocol 11.g reach a maximum of about 20 Mbit/s^{[citation required]}. The 11.n protocol has been used with the capacity to transmit 600 Mbit/s. The physical layer supports a rate of 300 Mbit/s, using two spatial streams within a 40 MHz channel. Depending on the environment, this can translate into a user-perceived throughput of 100 Mbit/s.