Richard,

Yes I am referring eBGP scenario. I suggest distance calculations based on

1. ICMP delay between eBGP speakers.
2. Manually configure  binding to remote AS.


Each eBGP speaking routers need to accumulate  distance value when advertised 
routes to external peer.There is no reqirment to synchronize different 
administrative domains since router itself automatically calculate value and 
add when routes  advertised similar to AS PATH addition operation.





----- Reply message -----
From: "Richard Li" <renwei...@huawei.com>
To: "Duleep Thilakarathne" <dule...@mobitel.lk>, "UTTARO, JAMES" 
<ju1...@att.com>, "'Robert Raszuk'" <rob...@raszuk.net>
Cc: "'bess@ietf.org'" <bess@ietf.org>
Subject: [bess] BGP route selection criteria - geographic distance when AS_PATH 
are equal
Date: Fri, Aug 7, 2015 10:11 PM

There might be a good point here. RFC 7311 only takes care of the IGP metrics. 
But In Duleep’s example, the metrics between two eBGP speakers are not taken 
into consideration. In order to have AIGP attribute to really represent the 
accumulated one, the metrics on such links should be considered as well. 
However, there might be some challenges or obstacles: The way to configure one 
metrics on the link between two eBGP speakers might not be consistent with the 
way to configure another metrics on the another link between two speakers.

Richard

From: BESS [mailto:bess-boun...@ietf.org] On Behalf Of Duleep Thilakarathne
Sent: Friday, August 07, 2015 8:41 AM
To: UTTARO, JAMES; 'Robert Raszuk'
Cc: 'bess@ietf.org'
Subject: Re: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Jim,


What I want to suggest is to insert item 5 (refer below items listed) to BGP 
best path selection algorithm. Once AS-PATH length is equal, next we can think 
on how to select best outgoing interface. If we don’t select proper outgoing 
interface it will affect to latency. I am talking this based on practical 
experience I have in ISP environment. There are several options to select best 
outgoing interface when AS-PATH are equal.  In this case I suggest geo distance 
to destination. Following are options to calculate geo distance. Router selects 
outgoing interface with lowest GEO distance to destination.

1. BGP speaking router can add distance when advertise to route to upstream 
similar to AS-PATH attribute. For example

A----B----C-----D

Router B advertise distance AB to router C. router C advertise accumulated 
distance AB+BC to router D.

2. Above distance can be configured as manual interface command or dynamically 
using ICMP or similar mechanism. We can assume ICMP delay propositional to geo 
distance.

3. Alternative option is to calculate real geo distance from coordinate system. 
In this case we miss intermediate hops. Accuracy is not much accurate since 
cable paths do not follow real coordinate based distance. In this case we 
should have knowledge on coordinates of upstream router which relevant IP block 
advertise.

Option 1 can be achieved through BGP protocol itself ,if agreed to introduce 
new attribute.
Option 3 more suitable to SDN based implementation. Calculation can be daily or 
weekly basis as this is not primary criteria.

Further Consider following scenario.

I am in Sri Lanka. Assume I have upstream POPS to Singapore, AMS, New York.  
Assume I need to reach destination IP located at Japan. When I check BGP 
routing table, AS-PATH length is equal from all three upstream. Then I have 
three options. Then router selects any interface randomly if no policy 
configured. I hope you agreed up to this point. In such case I suggest to 
consider GEO distance to destination. In most cases lowest distance path is the 
best path. This may not correct always but better than random outgoing 
interface selection.




1. Discarding the routes with the unreachable Next_Hop.






2. Preferring the route with the highest Local_Pref.






3. Preferring the aggregated route. The preference of an aggregated route is 
higher than the preference of a non-aggregated route.






4. Preferring the route with the shortest AS-Path.






5. If AS-Path finds equal, consider shortest GEO distance. If still distance is 
same follow next steps.






6. Comparing the Origin attribute and selecting the routes with the Origin 
attribute as IGP, EGP, or Incomplete in order.




Regards
Duleept



From: UTTARO, JAMES [mailto:ju1...@att.com]
Sent: Friday, August 7, 2015 7:44 PM
To: Duleep Thilakarathne; 'Robert Raszuk'
Cc: 'bess@ietf.org'
Subject: RE: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Duleep,

                So a bit confused here.

                How do want the decision making to go if a path has a shorter 
AS-PATH and longer latency than the alternative?? If latency is the prime 
motivator why do you care about AS-PATH length at all.. Comments In-Line..

Jim Uttaro

From: Duleep Thilakarathne [mailto:dule...@mobitel.lk]
Sent: Friday, August 07, 2015 9:31 AM
To: Robert Raszuk
Cc: UTTARO, JAMES; bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Hi Raszuk,


Question 1: How does the router know about user's high latency ?

Actually I am referring ISP edge router to another ISP edge router delay due to 
transmission distance.
[Jim U>] The underlying facility and it’s representative transmission distance 
will most likely differ from geographical distance. Which do you want to 
address? To Robert’s point you still need to acquire that knowledge and it may 
be orthogonal to an attribute that is defined as delay.


Question 2: How do you assure Internet stability where you start churning paths 
based on the latency of data plane ?

It is not required to consider stability in this situation since it is 
unavoidable. What is refer is, router need to select best outgoing path 
considering physical distance whenever possible when AS-PATH length is equal. 
If router selects long distance path randomly, it impacts to latency.

Question 3: What you are after has effectively been solved many years ago .. it 
is called Optimized Edge Routing (OER) / Performance Routing (PFR) - I suggest 
you google for those terms.

Thank for the suggestion. I gone through these proposals. But what I am 
suggesting is  whether we can address this idea from BGP protocol level. For 
example by introducing new attribute related to physical distance/delay similar 
to AS-PATH. New attribute need to update across the As path. My ultimate 
objective is to prevent router randomly select outgoing path when AS-PATH 
lengths are  equal. Further I am trying SDN based simulation these days. Hope I 
can share output. But this could similar to what you have proposed except geo 
distance calculation mechanism.

Refer below standard BGP route selection criteria. I suggest item 5. Wordings 
may different from vendor to vendor.




1. Discarding the routes with the unreachable Next_Hop.






2. Preferring the route with the highest Local_Pref.






3. Preferring the aggregated route. The preference of an aggregated route is 
higher than the preference of a non-aggregated route.






4. Preferring the route with the shortest AS-Path.






5. If AS-Path finds equal, consider shortest GEO distance. If still distance is 
same follow next steps.






6. Comparing the Origin attribute and selecting the routes with the Origin 
attribute as IGP, EGP, or Incomplete in order.








Regsrds
Duleept

From: rras...@gmail.com<mailto:rras...@gmail.com> [mailto:rras...@gmail.com] On 
Behalf Of Robert Raszuk
Sent: Friday, August 7, 2015 6:29 PM
To: Duleep Thilakarathne
Cc: UTTARO, JAMES; bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Duleep,

> Then end user experiences high latency to reach destination. In such
> a case, I suggest router need to consider geographic distance to
> destination and select path via NTT to reach destination by default.

Question 1: How does the router know about user's high latency ?

Question 2: How do you assure Internet stability where you start churning paths 
based on the latency of data plane ?

Question 3: What you are after has effectively been solved many years ago .. it 
is called Optimized Edge Routing (OER) / Performace Routing (PFR) - I suggest 
you google for those terms.

Regards,
R.












On Fri, Aug 7, 2015 at 2:51 PM, Duleep Thilakarathne 
<dule...@mobitel.lk<mailto:dule...@mobitel.lk>> wrote:
Hi Jim,

Please refer below example.

Assume destination IP is in Asian region. Particular ISP in a different  
location (Say India) has upstream peering to US POP (Say AT&T) and Asia POP 
(Say NTT). If we check BGP routing table, assume it shows

XX.XX.XX.XX/24 -------->AS - AT&T,AS-XX,AS-Destination
                                -------->AS - NTT,AS-YY,AS-Destination


In above case AS-PATH is equal and assume router automatically select path via 
AT&T. Then end user experiences high latency to reach destination. In such a 
case, I suggest router need to consider geographic distance to destination and 
select path via NTT to reach destination by default. Deciding geo distance is a 
challenge but there are options. Here geo distance means shortest distance to 
reach IP destination from upstream POP. Current practice is to use community 
strings, but it depends on upstream ISP capability.

Can you comment my idea.

Regards
Duleept



From: UTTARO, JAMES [mailto:ju1...@att.com<mailto:ju1...@att.com>]
Sent: Friday, August 7, 2015 4:09 PM
To: Duleep Thilakarathne; 'Robert Raszuk'

Cc: 'bess@ietf.org<mailto:bess@ietf.org>'
Subject: Re: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Duleep,

                Assuming AS-PATH is equal and AS-Content different how can you 
know that the internal metrics of each AS are consistent and mirror actual 
geographic distances? You have to be assured that each administrative domain 
applies the same metric assignment. I do not believe this is possible when 
there are multiple administrative domains.

Jim Uttaro

From: BESS [mailto:bess-boun...@ietf.org] On Behalf Of Duleep Thilakarathne
Sent: Friday, August 07, 2015 5:19 AM
To: Robert Raszuk
Cc: bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Hi Raszuk,

I went through RFC7311 and my concern is different than RFC 7311. I have 
analyzed full BGP routing table (541,199 routes) with two tier 1 ISP 
multi-homing scenario and found nearly 50% of routes have equal AS-PATH length. 
In this analysis It was considered, there was no route policy applied to 
influence local preference. According to BGP best path selection algorithm, 
when AS-PATH lengths  are equal, router breaks tie condition based on route 
internal logic. This does not grantee proper outgoing path selection.

Appreciate your concern on above analysis.

Regards
Duleept







From: Robert Raszuk [mailto:rob...@raszuk.net]
Sent: Monday, July 27, 2015 2:40 AM
To: Duleep Thilakarathne
Cc: bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] BGP route selection criteria - geographic distance when 
AS_PATH are equal

Hi Duleep,

Please consider RFC 7311 and provide feedback why you think it is not 
sufficient for your objective.

https://tools.ietf.org/html/rfc7311

Best,
R.


On Sun, Jul 26, 2015 at 9:15 PM, Duleep Thilakarathne 
<dule...@mobitel.lk<mailto:dule...@mobitel.lk>> wrote:
Hi,

I would like to suggest to consider geographic distance when AS_PATH  are equal 
in BGP route selection criteria. (as tie breaking rule). Can anybody comment on 
my idea.


Regards
Duleept




This e-mail and any attachments may contain confidential and
privileged information. If you are not the intended recipient,
please notify the sender immediately by return e-mail, delete this
e-mail and destroy any copies. Any dissemination or use of this
information by a person other than the intended recipient is
unauthorized and may be illegal.
Mobitel (Pvt) Ltd.

_______________________________________________
BESS mailing list
BESS@ietf.org<mailto:BESS@ietf.org>
https://www.ietf.org/mailman/listinfo/bess

This e-mail and any attachments may contain confidential and privileged 
information. If you are not the intended recipient, please notify the sender 
immediately by return e-mail, delete this e-mail and destroy any copies. Any 
dissemination or use of this information by a person other than the intended 
recipient is unauthorized and may be illegal. Mobitel (Pvt) Ltd.
This e-mail and any attachments may contain confidential and privileged 
information. If you are not the intended recipient, please notify the sender 
immediately by return e-mail, delete this e-mail and destroy any copies. Any 
dissemination or use of this information by a person other than the intended 
recipient is unauthorized and may be illegal. Mobitel (Pvt) Ltd.

This e-mail and any attachments may contain confidential and privileged 
information. If you are not the intended recipient, please notify the sender 
immediately by return e-mail, delete this e-mail and destroy any copies. Any 
dissemination or use of this information by a person other than the intended 
recipient is unauthorized and may be illegal. Mobitel (Pvt) Ltd.
This e-mail and any attachments may contain confidential and privileged 
information. If you are not the intended recipient, please notify the sender 
immediately by return e-mail, delete this e-mail and destroy any copies. Any 
dissemination or use of this information by a person other than the intended 
recipient is unauthorized and may be illegal. Mobitel (Pvt) Ltd.


This e-mail and any attachments may contain confidential and
privileged information. If you are not the intended recipient,
please notify the sender immediately by return e-mail, delete this
e-mail and destroy any copies. Any dissemination or use of this
information by a person other than the intended recipient is
unauthorized and may be illegal.
Mobitel (Pvt) Ltd.
_______________________________________________
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess

Reply via email to