|
North American Network Operators Group Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical Re: Anycast 101
On 17-dec-04, at 15:27, William Allen Simpson wrote: In short: a customer must pplb across two routers at the same ISP, and each of those routers must have different preferred paths to different anycast instances. This isn't going to happen often, but it's not impossible, and it's not bad engineering on the customer's or ISP's part if it does, IMO. You're wrong. That's VERY bad engineering! It doesn't really require that. Redundancy requires that the routers at the ends of two links both be different. Having one router at one end and two at the other is a good compromise in many situations.PPLB requires 2 routers, one at each end of the link bundle. How many people run multicast exactly? And that's precisely the reason why multicast is a different SAFI so you get to have different multicast and unicast routing.More than 1 router at any end will lead to a lot more problems than anycast, including multicast and any stateful protocol (like TCP). As for TCP, it would be very useful if someone were to run the following experiment: +-------+ |router2| +-------+ / \ +------+ +-------+ +-------+ +------+ |host a+---+router1| |router4+---+host b| +------+ +-------+ +-------+ +------+ \ / +-------+ |router3| +-------+ (Assuming the links to the hosts are (for instance) gigabit and the ones between the routers fast ethernet. If they're all the same speed you're only going to see out of order packets when the later packet is smaller than the earlier packet, which is inconsistent with a TCP session running at full blast.) Setup #1: per destination load balancing from router1 to routers 2 and 3 Setup #2: per packet load balancing from router1 to routers 2 and 3 So setup #1 will get no reordering, but is limited to 100 Mbps. Setup #2 will see reordering, but has a total bandwidth of 200 Mbps. Which going to perform better? Traffic patterns aren't the same in both directions in many/most cases, so one direction may be enough.For one thing, the load balancing will be only in 1 direction, and will lead to congestion in the reverse path.... Self defeating.
|