Yes, I would like to expand the topics that I am covering to generally deal with much of the CCNP ENCOR exam. Once I'm finished wth the fundamentals of BGP, I'll go on to OSPF and some other topics.
Great explanation as always . I have two questions if you can answer will be a great favour . 1) Other than the scalability factor what other main benefit Route Reflectors carries . Like in full mesh each router need to have ibgp peership and will exchange routes thus increasing configs and processing issues. But the same thing RR is doing being alone getting routes and reflecting etc ? 2) Where will be the convergence fast ? In full mesh or RR and why ?
Hello @afaqshah7229! I'm glad you you find the explanation helpful! Concerning your first question, the main benefit is indeed scalability, but when we talk about scalability, we are talking about several different factors that affect it. These factors include: 1) Reduced configuration complexity - With RRs, only clients need to maintain sessions so you have a significantly smaller number of iBGP sessions to configure (remember, BGP sessions do not form automatically like in OSPF or EIGRP, they must be manually configured). 2) Lower resource utilization, except of course for the RR. In this case, you simply have to provision for more robust (and expensive) routers to play the role of the RRs. You're right that the RR will still have a large operational load. But the client routers won't need to have such large and expensive resources (memory, CPU power etc). 3) Simplified network management - It's easier to manage and troubleshoot a network with RRs rather than with a full mesh iBGP network. There are fewer BGP sessions to monitor, and route propagation is more predictable. 4) You can centralize routing policy applications by applying them at RRs ensuring consistent routing behavior across the network, rather than having to configure the policies in every router. So you see, it is really about scalability, but there are various aspects of scalability that RRs address. Does that make sense? Concerning your second question, there are several factors that will affect convergence directly and indirectly. Generally speaking, a full mesh iBGP peering topology will converge slightly faster than a well designed RR topology. However, if the number of routers are extensively large (i.e. on the order of 50, 100 or more), then you will begin to tax resources such as CPU, memory, as well as bandwidth. If those resources are not enough to maintain timely communication between BGP routers, then convergence will obviously slow down. So after a certain size, a full mesh topology will tax the resources necessary to make it converge fast, thus slowing it down. Ultimately, even if an RR topology converges somewhat slower, the benefits delivered by RRs are still preferrable. Remember, BGP is not all about fast convergence. It is about stability. I hope that answers your questions!
@@Lazarus.telecomTech Thank You very much sir for the very detailed explanation . Just one question if you can briefly explain . As you said Full Mesh can converge fast if few routers are there . What is the main reason for fast convergence ?
Hello@@afaqshah7229 ! The primary reason for which a full mesh will converge faster than a route reflector topology is because you have direct connections between iBGP peers. In a full mesh iBGP configuration, every router is directly connected to every other router within the AS. This means that routing updates are sent directly to every router without intermediary steps. Since each router receives routing updates directly from every other router, the propagation of routing information is faster. There are no delays caused by intermediate nodes processing and forwarding the updates. Each router processes and forwards updates simultaneously to all peers, minimizing the time taken for the entire network to be updated. Again, you always have to take into account the benefits and drawbacks of each solution based on your topology. Because BGP is not designed to be fast, but designed to be stable, you should choose the solution that will provide that stability in the best possible way.
@@Lazarus.telecomTech Great . Thank you very much for the detailed answer . In coming day if possible make a video for BGP Design . What are the key parameters to design BGP for a network .
Great tutorial as always. Cheers.
Thank you! Cheers!
thank you for the nice video!
Glad you liked it!
great lessons somany others topic we want u to covert like multicast-automation
Yep, I'll put those on my list as well!
Well done sir! Thanks!!
As always, thanks so much!!
let me be honest you are so great as a teatcher so please consider to give your knowledge of all CCNP Encor exam thx anyway
Yes, I would like to expand the topics that I am covering to generally deal with much of the CCNP ENCOR exam. Once I'm finished wth the fundamentals of BGP, I'll go on to OSPF and some other topics.
Great explanation as always . I have two questions if you can answer will be a great favour .
1) Other than the scalability factor what other main benefit Route Reflectors carries . Like in full mesh each router need to have ibgp peership and will exchange routes thus increasing configs and processing issues. But the same thing RR is doing being alone getting routes and reflecting etc ?
2) Where will be the convergence fast ? In full mesh or RR and why ?
Hello @afaqshah7229! I'm glad you you find the explanation helpful!
Concerning your first question, the main benefit is indeed scalability, but when we talk about scalability, we are talking about several different factors that affect it. These factors include:
1) Reduced configuration complexity - With RRs, only clients need to maintain sessions so you have a significantly smaller number of iBGP sessions to configure (remember, BGP sessions do not form automatically like in OSPF or EIGRP, they must be manually configured).
2) Lower resource utilization, except of course for the RR. In this case, you simply have to provision for more robust (and expensive) routers to play the role of the RRs. You're right that the RR will still have a large operational load. But the client routers won't need to have such large and expensive resources (memory, CPU power etc).
3) Simplified network management - It's easier to manage and troubleshoot a network with RRs rather than with a full mesh iBGP network. There are fewer BGP sessions to monitor, and route propagation is more predictable.
4) You can centralize routing policy applications by applying them at RRs ensuring consistent routing behavior across the network, rather than having to configure the policies in every router.
So you see, it is really about scalability, but there are various aspects of scalability that RRs address. Does that make sense?
Concerning your second question, there are several factors that will affect convergence directly and indirectly. Generally speaking, a full mesh iBGP peering topology will converge slightly faster than a well designed RR topology. However, if the number of routers are extensively large (i.e. on the order of 50, 100 or more), then you will begin to tax resources such as CPU, memory, as well as bandwidth. If those resources are not enough to maintain timely communication between BGP routers, then convergence will obviously slow down. So after a certain size, a full mesh topology will tax the resources necessary to make it converge fast, thus slowing it down.
Ultimately, even if an RR topology converges somewhat slower, the benefits delivered by RRs are still preferrable. Remember, BGP is not all about fast convergence. It is about stability.
I hope that answers your questions!
@@Lazarus.telecomTech Thank You very much sir for the very detailed explanation .
Just one question if you can briefly explain . As you said Full Mesh can converge fast if few routers are there . What is the main reason for fast convergence ?
Hello@@afaqshah7229 ! The primary reason for which a full mesh will converge faster than a route reflector topology is because you have direct connections between iBGP peers. In a full mesh iBGP configuration, every router is directly connected to every other router within the AS. This means that routing updates are sent directly to every router without intermediary steps. Since each router receives routing updates directly from every other router, the propagation of routing information is faster. There are no delays caused by intermediate nodes processing and forwarding the updates. Each router processes and forwards updates simultaneously to all peers, minimizing the time taken for the entire network to be updated.
Again, you always have to take into account the benefits and drawbacks of each solution based on your topology. Because BGP is not designed to be fast, but designed to be stable, you should choose the solution that will provide that stability in the best possible way.
@@Lazarus.telecomTech Great . Thank you very much for the detailed answer . In coming day if possible make a video for BGP Design . What are the key parameters to design BGP for a network .