Slashdot Mirror
Does Anyone Still Use Token Ring?
blanchae asks: "Does anyone still use Token Ring, or is it dead? I remember hearing about 100 mbps TR a few years ago but nothing since. I remember that the strong point of TR over Ethernet was the QOS and the consistent response time. Does the banking community still use TR?"
You're thinking of HSTR - and no, no one really uses it anymore. In looking around I was amazed to see that the working group even thought far enough ahead to start planning a gigabit spec. I havent seen a concentrator/MAU (right word?) in years, though. Any QoS features that were implemented in Token Ring are pretty much duplicated in 802.1p and other (proprietary) layer 2 QoS/CoS protocols.
Rings themselves are still used, just in other topologies. You may still see some FDDI here and there, and many cable companies use RPR/DTP/SRP to deliver digital cable and broadband access at the same time in their cores.
Either way, I'm sure the pointy haired boss doesn't miss it.
There is clearly a lot of research going on, with results published, as in Raja's Optimal bandwidth utilization in wireless token ring networks released earlier this year. However, 1998 was the last big year for user's guides, which indicates that this technology has long since fallen from the mainstream and now survives only in academia.
Nope, banks use ip over ethernet, and atm over fiber. Backbone to ATM (well remote atms can be dialup, isdn, or frame), its all modern (or reasonably so).
I've seen token ring in some banks but only old jacks(I work in the industry).
The only place I've seen it used still is in the retail (POS) market, IBM calls it store loop though, and its even slower(kbps IIRC). Most of those places are scraping to get rid of it, not nessisarily for any reason other than its easy to drop the whole network(wires are aged, stupid clerks, etc.), and expensive to negate that problem with one of those relay doo-dads.
Mostly, IMHO, anyone that still uses it wants it gone. Its dead jim.
From [[Token ring]]:
Madge Networks, a one time competitor to IBM, is now considered to be the market leader in Token Ring.
From [[Madge Networks]]:
Madge Networks NV. was a global leader and pioneer of high speed networking solutions in the mid 1990s. The company was founded by Robert Madge.
The company filed for bankruptcy in April 2003.
Granted, they still exist, and sell stuff, but for a market monopoly to file for bankruptcy...can't be too many customers left, can there?
The big deal with token ring was that the network would remain stable under 100% load. Classic 10mbps ethernet with hubs would start experiencing trouble around 60% load and collapse by the time load reached 90%. If you had a big, flat network it just plain wouldn't work.
Look at why: With token ring, only the card holding the token could transmit. Everybody else had to wait for the token. So each station would empty its transmit queue and then pass the token on to the next station. On ethernet, a station would send a packet whenever and if another station sent a packet at about the same time they'd collide. Every station observing the collision would assert a collision signal and after the collision signal cleared the two stations that transmitted would wait a random period of time and then retransmit. That's oversimplifying a bit but more or less correct.
So, token ring was much more stable in a large LAN with a high probability of multiple stations having outbound traffic ready at the same time.
Now, along comes 100baseTX on cat5, the end of coaxial ethernet and the proliferation of $50 switches. When you're plugged in to a switch there are only two devices in the collision domain: you and the switch. So, lots less collisions. When you're in full duplex mode (as you generally are), collisions are impossible since by definition both sides are allowed to transmit at the same time. Now your ethernet network remains stable at 100% utilization. And if the nic in the PC burns out, the rest of the network doesn't care.
Token ring is very sensitive to malfunctioning nics. A malfunctioning nic may drop the token, that is it may receive the token and then fail to transmit it to the next nic. That kills a token ring network dead until the admin wanders around with an analyzer and figures out which PC is at fault.
Suddenly the tables were turned. Token ring was an administrative headache and expensive to boot. Ethernet was simple, cheap and worked just as well.
Token ring died out except as an academic curiosity -- an interesting early answer to a problem that was eventually solved another way.
Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion.
We suffered through that just long enough to realize that it's completely non-maintainable. We ended up turning their ring topology into a sort-of star by buying devices called Autoshunts, which were banks of relays and detector circuits. We wired each register all the way back to the Autoshunt, rather than daisy chaining them. When a cash register went offline, the Autoshunt would drop that register's relay, preserving the loop for the other registers. (When token ring came around, the MAU performed a similar task.)
It was a clever hack. The Autoshunts did a good job keeping most registers on line, and reporting to the store person any offlines. But they were expensive, and with so many moving parts they failed often enough on their own. And they were fickle -- sometimes they'd recognize the offline token, sometimes they wouldn't.
In reality, the ring topology is a total pain to maintain. Self-healing technology makes it mostly workable, but it really adds a larger layer of complexity than you might imagine. The modern implementation of ethernet via star (hubs/switches) is far more flexible, and far simpler to maintain.
John
I've been working as a sysadmin for the last 8 years. Everybody uses ethernet these days, but there are exceptions. Some big networks still have segments that are token-ring, usually in older buildings. I even know a company that still uses token-ring because upgrading to ethernet would mean recabling the whole 8-story building. And their network apps work fine over their token-ring network. If it works, don't fix it.
>And we did get the 100Mb token ring switches, which was truly one of the more absurd things I have ever seen IT money spent on. I still don't have a clear idea how this was a good thing: you got a 100Mb token ring switch, which would create a ring on each port. Then you could plug exactly one device into each port, as long as it had a 100Mb token ring adapter.
As you say, it's a bit absurd to use expensive TR switches like that (instead of cheap Ethernet switches) -- since in a single-NIC-per-port arrangement there's no chance for collisions in any case, so TR's main advantage is meaningless.
Still, it made sense to migrate to TR switches -- but by having small rings of clients share switch ports, and dedicating a port to a single system only for real bottleneck systems such as file servers. If you ask me, the real reason to stick with TR would have been that switching to Ethernet would meant either replacing everything at once (prohibitive in labour cost and downtime) or a potentially messy gradual transition either with routers (and a whole lot of reconfiguration of systems) or translational bridging (here be dragons).
Everyone is moving to wifi, anyway.
NO!! That's way to general of an assumption and is just wrong. It's like saying everyone is deploying CAT5e, or everyone is deploying fiber. I would say it is a mix of the three, cat5e for Ethernet (up to 10Gbe soon!), fiber for higher speed or longer run needs, and wifi where needed. How ever, we have setup a lot of networks and the vast majority of people (businesses any way) and still deploying CAT5e. Wireless is just nerver going to be fast enough, secure enough, and reliable enough to push copper and fiber out of the picture. So no, everyone isn't moving to wifi, they are deploying it along side copper and fiber but in a more limitied capacity and only where nessecary. Wired links are still better in most cases and this is what we promote to customers.
I have a laptop...I'd have to find a PCMCIA TR card.
I have one, made by IBM. I'd sell it.
Not true. There are QOS issues that Ethernet cannot possibly guarantee. With token ring, on a fully loaded network, you are GUARANTEED to receive the token and therefore be able to send data with a certain frequency. Therefore, if you have data that MUST be sent and received with certain maximum delay characteristics, Token Ring wins hands down. Ethernet simply cannot guarantee anything of the sort. Switches reduce collisions, but if there is an upstream link in the design, you can still overflow the buffers if you have enough data to overload that link's capacity. Admittedly, these are extremely rare requirements, but they do arise. Also, as somebody had mentioned in a previous post, FDDI is an advanced version of Token Ring, with dual rings (running opposite directions) in the event that one segment is cut. The second ring passes data backwards to the last node before the cut, therefore replacing the cut segment at the cost of a slight delay increase, but as traffic on this ring is not processed by any node except the last, delay is extremely minimal.