I have conducted the similar testing using 802.11b/g chipset (ath5k). Previous testing is done based on 802.11n chipset (ath9k). The TCP throughput is able to achieve 22Mbps (maximum achievable throughput) from Mesh A1 to Mesh B2. I have done it several time and the results are consistent.
Maybe we need to check whether it is caused by integration between ath9k and mesh 802.11s. Can someone comment on this? On Wed, Apr 20, 2011 at 1:22 PM, Yeoh Chun Yeow <yeohchuny...@gmail.com>wrote: > I have reboot the Mesh A2 bridge Mesh B1. Then, I can manage to achieve > TCP throughput 17.8Mbps from Mesh A1 to Mesh B2 and tx retries are reduced. > It's weird. > > > On Wed, Apr 20, 2011 at 1:01 PM, Yeoh Chun Yeow <yeohchuny...@gmail.com>wrote: > >> >>> - If you are planning on deploying a multi channel mesh, it will pay >>> off very quickly to extend open80211s to support it and not do >>> bridging. Bridging is intended for heterogeneous intefaces (e.g. mesh >>> and AP, or mesh and ethernet). But it adds additional overhead (an >>> additional address lookup per frame, larger mesh headers, etc.). >>> >> >> Please point to the relevant document on how to implement the extension of >> open80211s to support multi channel mesh. >> >> >>> To further investigate the lower throughput you observe, you should look >>> at sniffer captures. >> >> Are mesh frames transmitted at the expected data rate? >> >> Sniffer captures shows that the mesh stations are using 54Mbps data rate >> (observed in Radiotap header). >> >> How busy is the channel? >> >> The channel should be clean and not much noise. Noise level are -115dBm >> (HP card) and -95dBm based on survey dump. Signal are -55dBm amd -47dBm >> respectively. >> >> >>> Are there many retries/losses? >> >> Mesh A1 < ----- Channel 116 -------> Mesh A2 bridge Mesh B1 <----- Channel >> 40 ------> Mesh B2 >> >> TCP throughput from Mesh A1 to Mesh A2 is 17.8Mbps, no tx failed on Mesh >> A1, but tx retries 795. >> >> TCP throughput from Mesh A1 to Mesh B2 is 9.57Mbps, no tx failed on Mesh >> A1, but tx retries 138. Interesting here is Mesh B1 records no tx failed, >> but tx retries is 4932. This could be the root cause of the problem. Too >> many retries in Mesh B1 to Mesh B2. But if you measure throughput from Mesh >> B1 to Mesh B2, you manage to get 17.5Mbps, with no tx failed and tx retries >> 774. >> >> >
_______________________________________________ Devel mailing list Devel@lists.open80211s.org http://open80211s.com/mailman/listinfo/devel
