ARCNET

ARCNET (Attached Resource Computer Network) is a networking technology for local area networks (LANs). It defines cable types and signaling for the physical layer and packet formats and protocols for the media access control (MAC)/secure layer of the OSI model.

ARCNET was first introduced to the public on December 1, 1977, after Datapoint, the computer manufacturer that developed the technology, completed installation at its first customer, Chase Manhattan Bank in New York City, that same year. ARCNET was developed at the same time as Ethernet, but held as a proprietary stand-alone for several years until Standard Microsystems Corporation took over production of the corresponding controllers, achieved a release in 1982, and gained Novell as another partner in 1985. In the meantime, ARCNET has become one of the foundations of BACnet, for example.

ARCNET is physically structured either like a classic 10BASE2 Ethernet as a bus or alternatively as a star or tree. The switching components are simple T-pieces in the bus variant, or active or passive hubs in the star variant, whereby active and passive hubs can be mixed as desired. In its original form, ARCNET was built with coaxial cables. As development progressed, UTP and fiber optic were also specified. The coaxial cables do not correspond to the RG-58 type known from 10BASE2 (Thin Ethernet) (with a characteristic impedance of 50 ohms), but to RG-62 with a characteristic impedance of 93 ohms. This circumstance temporarily caused Arcnet to spread faster than Ethernet, since the RG-62 cables corresponded to IBM's SNA cabling and therefore in many places it was already possible to fall back on an existing infrastructure.

The connection to the hub and computer is optionally made in star form directly via BNC connectors without the T-piece known from 10BASE2 to the hub, whereby no terminating resistor is required, since the termination of the cables is made directly in the hub or on the card. The bus variant is wired in the same way as 10BASE2 with T-pieces and also has terminating resistors (93 Ohm) at the end of the cable. The possible distances between nodes are 2.5 to 4 times those of 10BASE2.

The ARCNET access method is Token Passing on a Token Bus. Similar to the Token Ring access method, a token is sent on its way and passed on in a defined sequence. With Token Ring, this is predetermined by the ring cabling. With ARCNET, the token is always sent throughout the network, and the next station in a logical sequence picks up the token. The sequence is given to the cards as node ID (consecutive numbering) on an externally accessible DIP switch; with current products, the ARCNET ID is communicated via the card's software.

The transmission rate of 2.5 Mbit/s (20 Mbit/s by using ARCNET-Plus cards) is lower than Ethernet and Token Ring. However, since, as with Token Ring, no collision slows down the speed of the transmission, a higher speed can already be achieved with 2.5 Mbit than with Ethernet networks, which are 4 times faster, especially with the highest network load. In 1990, the US network specialist Thomas Conrad also presented a (non-specified) 100 Mbit variant of ARCNET with RG-62 and fibre optic cabling under the name "TCNS", which had all the advantages of ARCNET over Ethernet and Token Ring. However, Compaq dropped this system after the acquisition of Thomas Conrad in favor of Fast Ethernet, which had become much cheaper in the meantime. In the early years, however, ARCNET was considerably cheaper than Ethernet or Token Ring in terms of components, cabling and maintenance costs.

The ARCNET cards were also a special feature. ARCNET was an open design and in particular also had a standardized card API, so that cards from different manufacturers could also work with a uniform driver. This practical simplification was later also used by many manufacturers with the widely used NE2000-compatible Ethernet cards. There, too, this reduced development costs and soon led to very inexpensive cards. Today's ARCNET cards, however, require manufacturer-specific drivers.

An ARCNET network with a hub in the middle gave a good example in explaining network topologies: the cabling formed a star, electrically the network was a bus and logically a ring. ARCNET has lost importance with the further spread of Fast Ethernet in local networks, but is still used in areas such as industrial manufacturing, printing technology, wind energy technology, but also medical technology and logistics.

References

  1. a b ARCNETworks (PDF; 140 kB) ARCNET Trade Association. Fall 1998. archived from the original on March 4, 2016. retrieved March 10, 2013.
  2. Craig Zacker, Paul Doyle et al, Understanding and Repairing Networks, Macmillan Computer Publishing, USA, 1996, Chapter 7.

Questions and Answers

Q: What is ARCNET?


A: ARCNET is a communications protocol used for local area networks.

Q: When did ARCNET first become widely available?


A: ARCNET first became widely available in the 1980s.

Q: What was ARCNET originally used for?


A: ARCNET was originally used for office automation tasks.

Q: Where is ARCNET still used today?


A: ARCNET is still used today in home automation and certain industrial applications.

Q: What type of topology does ARCNET use?


A: ARCNET uses a star-like topology.

Q: What type of mechanism does ARCNET use?


A: ARCNET uses a token-passing mechanism.

Q: How does ARCNET compare to other network protocols in terms of speed?


A: ARCNET is much slower than other network protocols, such as Ethernet, with a maximum speed of about 20 Mbit/s.

AlegsaOnline.com - 2020 / 2023 - License CC3