Routed Ring Topology

Hi All

can someone please explain what is a routed ring topology ?


also what are spanned VLANs ?
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Are you talking about a token ring network? Typically, any kind of a routing ring (loop) isn't a good thing.

Spanned vlan's are vlan's that are propagated to multiple devices.
virkshivrajAuthor Commented:
so you mean something like VTP which propagates VLANs across to client switches in VTP domain, is capable of spanning VLANs

by routed ring i mean,
in a scenario where there are 10 switches, all switches are Layer 3 capable, and they are physically in placed in different locations in a building. how would routed ring topology work in this scenario.
If I understand the question correctly, it is a design that I wouldn't necessarily recommend.  That being said, if you imagine your scenario (limited to 5 to make it simpler), where your connectivity is like this

     a - b - c - d - e - (back to same "a" at the beginning, to make a loop),

The fact that you brought up "Routed" means that it's layer 3, so then let's consider that your layer 3 addresses for the sites are as follows:

So..  To make this work, there would be a routing protocol working on the network (RIP, OSPF, etc), so that each switch (which is also a router in this case) knows about all the other networks, and knows the fastest way to get there.   Network "A" knows that to get to "B" or "E", it's directly attached, and can send directly, but to get to network "C", the shortest way is via B, and the way to get to D is via "E".

Now, let's say that someone cuts the cord between switch/router "B" and "C".  Users on network "A" can still get to network E, and router E will tell network "A" that it still has a route to network C, so network "A" will now send C traffic to E, to be forwarded along.  In fact, even those on B and C will still be able to communicate through the backup link by going  "the long way" around the loop.

This is all true and possible because it's a "routed network".  The Router (layer 3 boundary) is the device that sends traffic BETWEEN separately numbered networks.

NOW..  If you WANTED to, you could turn off all of the routing, and make this whole network a "Spanned VLAN", which is to say, you're just connecting one switch to another, with each switch having no care in the world about what IP addresses are on the network, and doing no routing.  In this "Layer 2" scenario, the switches do what they do best - They pass any packets they receive out to any and all ports on the switch that might have a computer that needs that packet on it.  They have a "Mac address table" that help them keep track of where different devices are, but if a packet is received for which there is no known destination, it will be "flooded" or sent out EVERY port on the switch.

So, imagine the first time this network gets turned on, and there's no knowledge of any machines on the network.  The first packet could go around and around the loop forever (I'm oversimplifying here - there is a  TTL mechanism, but I'm making a point) which would mean that ONE PACKET could take up all the bandwidth of all the switches and take down the network...

SO.  In this "switched network" or "Spanned VLan", there is a protocol called "Spanning tree" which all switches use.  You can read the details on that WiKi article, but the over-simple explanation is that every switch sends out occasional packets called BPDUs.  If a switch hears it's own BPDU in through another port, it knows there's a loop in the network that could cause problems.  All the switches negotiate, and one port that is in the path of the loop will automatically shut down.

Every time a new machine gets plugged into the network, a BPDU is sent to see if that port just created a new loop, and if it did, it will be dealt with.

This means that there will not be a loop in the network described above, and if the Spanning tree (also known as STP) shuts down the link between B and C (for example), then for B to talk to C, it has to go "the long way" through A, then E, then D, then finally to C.  

If someone then breaks the link between D and C, the spanning tree will notice that the BPDUs aren't making it over that link anymore, meaning the loop is gone, and it will renegotiate the whole thing, and turn the link between B and C back on to regain full connectivity.

There are pros and cons to every design.  Too many to discuss which may be best between these two without more information, so I wont try unless you have more questions.  I get the feeling this is more of an academic problem for you anyway, so let me know if you dont understand, and I'll try to draw a couple real pictures, or something..

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Determine the Perfect Price for Your IT Services

Do you wonder if your IT business is truly profitable or if you should raise your prices? Learn how to calculate your overhead burden with our free interactive tool and use it to determine the right price for your IT services. Download your free eBook now!

virkshivrajAuthor Commented:
thanks Steve

so the scenario I have is like this .. n its in a production environment.

2 Dlink DGS 3627 layer 3 switches in one single stack - core switch
4 dlink dgs 3427 layer 2 managed switches in one stack - server switches (esx hosts, netapp and other production server)
6 dlink dgs 3427 layer 2 managed switches in one stack - client switches

the design
proposed design
so clearly in this scenario I cannot use routed ring topology (the question was to clear out conceptual doubts)

in the design above black links are 10 Gbps copper links and red ones red ones are 10 Gbps fiber .. the fiber links are used to reach out for switches across building floors.

previously VRRP was implemented in the design, and for a small network like ours I think it is unnecessary. So I have proposed simplicity in managing our LAN switching. And this will give us more switch-port capacity on the core switch. The old design
existing design

one thing I don't understand is that why was forwarding BPUD disabled in the STP config settings on the switches. My understanding was that BPDU should be enabled in order for STP to function properly.

here is an example of STP settings from one of the switches  

STP Bridge Global Settings
STP Status            : Enabled
STP Version            : RSTP
Max Age                    : 20
Hello Timer            : 2
Forward Delay              : 15
Max Hops                    : 20
TX Hold Count              : 6
Forwarding BPDU      : Disabled
LBD Recover Time      : 60
NNI BPDU Address      : dot1ad


virkshivrajAuthor Commented:
any comments on my previous post .... any feedback will be appreciated
virkshivrajAuthor Commented:
good explanation of the concepts
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Networking Protocols

From novice to tech pro — start learning today.