[c-nsp] Performance issue on link

[CiscoNSP List]

Hi Arie - I've tested with no policing, and also policing egress...added ingress to see if it made any difference(It didnt).

I've changed to a shaping policy:

class-map match-any MATCH_ANY match any

policy-map 40MB_SHAPE
 class MATCH_ANY
  shape average 41943040

and added egress to each of the routers Ints, But still see poor tcp performance POP B -> POP A

Ill try some of the linux tuning mentioned to see if that helps.

Thanks again for your assistance.

From: avayner at cisco.com
To: cisconsp_list at hotmail.com; cisco-nsp at puck.nether.net
Subject: RE: [c-nsp] Performance issue on link
Date: Mon, 1 Apr 2013 23:52:10 +0000






Yes, you should be shaping on egress. 
Why are you policing on ingress?! 



Arie 







Sent via my mobile phone. 







-------- Original message --------

From: CiscoNSP List <cisconsp_list at hotmail.com> 

Date: 

To: "Arie Vayner (avayner)" <avayner at cisco.com>,cisco-nsp at puck.nether.net 

Subject: RE: [c-nsp] Performance issue on link 





Thanks Arie - I already have the following policy-maps on each router:







policy-map 40MBPS



 class class-default



    police 41943040 5242880 conform-action transmit  exceed-action drop 






Then on the L3 Ints between the POP's







service-policy input 40MBPS



service-policy output 40MBPS







Should I be shaping rather than policing?  







> From: avayner at cisco.com

> To: cisconsp_list at hotmail.com; cisco-nsp at puck.nether.net

> Subject: RE: [c-nsp] Performance issue on link

> Date: Mon, 1 Apr 2013 23:20:34 +0000

> 

> You need to shape your traffic to 40Mbps (or most likely slightly less than that).

> TCP bursts and your provider drops the traffic, which causes the TCP window to shrink.

> 

> Your switches cannot shape, so you would have to do it on the routers.

> 

> Basically, a classic sub-rate link problem:

> http://www.cisco.com/en/US/docs/solutions/Enterprise/WAN_and_MAN/QoS_SRND_40/QoSIntro_40.html#wp61026

> 

> Arie

> 

> -----Original Message-----

> From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of CiscoNSP List

> Sent: Monday, April 01, 2013 15:51

> To: cisco-nsp at puck.nether.net

> Subject: [c-nsp] Performance issue on link

> 

> Hi,

> 

> We have a 40Mb link between 2 POPs - Latency ~65m/sec (No packet-loss)

> 

> POP A Is a 7301 and 2960POP B is a 7200 and 4948

> 

> 40Mb link is connected to the two switches (L2), and then a trunk link to both routers for all L3.

> 

> Have a Linux server connected to both switches, and achieve the following performance:

> 

> IPERF (UDP)

> POP A -> POP B - 38.5Mb/secPOP B -> POP A - 38.5Mb/sec

> 

> IPERF (TCP)

> POP A -> POP B - ~20Mb/secPOP B -> POP A - ~12Mb/sec

> 

> FTP

> POP A -> POP B - ~38Mb/secPOP B -> POP A - ~16Mb/sec

> 

> WGET

> POP A -> POP B - ~30Mb/sec POP B -> POP A - ~16Mb/sec

> 

> Any suggestions on why I am seeing poor performance with TCP transfers? (Especially POP B -> POP A direction) - I've tried adjusting the window size in IPERF but it actually made the results worse?

> Thanks in advance.

> 

> 

> 

> 

> _______________________________________________

> cisco-nsp mailing list cisco-nsp at puck.nether.net https://puck.nether.net/mailman/listinfo/cisco-nsp

> archive at http://puck.nether.net/pipermail/cisco-nsp/