Happy 50th birthday, Ethernet [Long article: > 5,000 words]

[6Foot2]

Link: Happy 50th birthday, Ethernet | APNIC Blog

 

Some 50 years ago, at the Palo Alto Research Centre (PARC) of renowned photocopier company Xerox, a revolutionary approach to local digital networks was born. On 22 May 1973, Bob Metcalf authored a memo that described ‘X-Wire’, a 3Mbps common bus office network system.

 

There are very few networking technologies from the early 70s that have proven to be so resilient (TCP/IP is the only other major networking technology from that era that I can recall), so it’s worth looking at Ethernet a little closer in order to see why it has enjoyed such an unusual longevity.

 

Hang on, was that 3Mbps? True, the initial Ethernet specification passed data at a rate of 2.94Mbps, not 10Mbps. That came later, as did the subsequent refinements to the Ethernet technology that allowed data transfer rates of 100Mbps, 1Gbps, and then 100Gbps. Terabit Ethernet (TbE) is on the way, by all accounts.

 

Why has Ethernet endured as an ubiquitous data framing protocol over all these years? What was the difference between Ethernet and a variety of other emerging technologies for local area networking, such as Token Ring, Fiber Distributed Data Interface (FDDI), or Distributed-queue dual-bus (DQDB)? Why didn’t we adopt simple HDLC serial framing? What’s so special about Ethernet?

 

The first question to ask is — has Ethernet really endured? Well, not really. You couldn’t take a piece of 75-ohm coaxial cable from the 1970s, increase the speed of the clocks in the electronics and run a common bus local area network using a data clocking rate of 200Gps. That just won’t work.

 

So, today’s high-speed Ethernet networks have some bits in common with the earlier Ethernet networks, and some are radically different. However, it is still possible to take an Ethernet packet frame off a high-speed Ethernet network and pass it into a 10Mbps Ethernet LAN. The Ethernet packet framing protocol and the station addressing plan are both largely unchanged.

 

Perhaps we should start this look at Ethernet from the perspective of 50 years ago and look at why Ethernet was so interesting for computer communications...