Skip to contentSkip to conversation

TCP stream performance regression due to c377411f2494a931ff7facdbb3a6839b1266bcf6

DISCUSSION Alex,Shi 3 participants 3 messages Jun 18, 2010 – Jun 24, 2010

lore
Share & tools
lore

Alex,Shi

author
#0 lore raw

In our netperf testing, TCP_STREAM56 shows about 20% or more performance regression on WSM/NHM and tigerton machines. The testing boot up both netserver and client on localhost. The testing command like this:
./snapshot_script_net TC_STREAM56 127.0.0.1

We found the following commit causes this issue. c377411f2494a931ff7facdbb3a6839b1266bcf6
Revert this commit will recover this regression on all machine.

Regards!
Alex

#1 lore raw

More info about the testing:
It's a loopback testing. We start one client netperf process to communicate with netserver process in a stream TCP testing. To reduce the cpu cache effect, we bind the 2 processes on 2 different physical cpus. #taskset -c 0 ./netserver
#taskset -c 15 ./netperf -t TCP_STREAM -l 60 -H 127.0.0.1 -i 50,3 -I 99,5 -- -s 57344 -S 57344 -m 4096

Quoted 13 lines
Quoted 13 lines

-----Original Message-----
From: Shi, Alex
Sent: 2010年6月18日 15:17
To: [email protected]; [email protected]
Cc: Chen, Tim C; Zhang, Yanmin; [email protected]
Subject: TCP stream performance regression due to c377411f2494a931ff7facdbb3a6839b1266bcf6

In our netperf testing, TCP_STREAM56 shows about 20% or more performance regression on WSM/NHM and tigerton machines. The testing boot up both netserver and client on localhost. The testing command like this:
./snapshot_script_net TC_STREAM56 127.0.0.1

Quoted 6 lines
Quoted 6 lines

We found the following commit causes this issue. c377411f2494a931ff7facdbb3a6839b1266bcf6
Revert this commit will recover this regression on all machine.

Regards!
Alex

�翳簕.n�+壏煯壄�+%娝�遍荻w�簕.n�+壏{炳G珴�妠ay�蕠跈�,jf"穐殢飦�戧鐉_璁(殠娸"濟�m���G珴�⒏?櫒璀�&x忈秈O曟瑉窔v豝�m��� �鹅�⒏?朓�

#2 lore raw
Quoted 5 lines · Yanmin

More info about the testing:
It's a loopback testing. We start one client netperf process to communicate with netserver process in a stream TCP testing. To reduce the cpu cache effect, we bind the 2 processes on 2 different physical cpus. #taskset -c 0 ./netserver
#taskset -c 15 ./netperf -t TCP_STREAM -l 60 -H 127.0.0.1 -i 50,3 -I 99,5 -- -s 57344 -S 57344 -m 4096

Thanks guys

We corrected a proven vulnerability in network stack.

If you want better netperf results, just increase size of your sockets buffers. 57344 is _very_ low for localhost communication, given lo MTU is 16436.

As we also increased skb->truesize lately, you might also increase buffer size regardless of this (net: sk_add_backlog() take rmem_alloc into account) commit.

In my case, I saw following improvement :

Under huge stress from a multiqueue/RPS enabled NIC, a single flow udp receiver can now process ~200.000 pps (instead of ~100 pps before the patch) on a 8 core machine.

Thats a 200.000 % increase, in a situation no tuning was possible ;)

Quoted 20 lines
Quoted 19 lines
Quoted 19 lines

-----Original Message-----
From: Shi, Alex
Sent: 2010年6月18日 15:17
To: [email protected]; [email protected]
Cc: Chen, Tim C; Zhang, Yanmin; [email protected]
Subject: TCP stream performance regression due to c377411f2494a931ff7facdbb3a6839b1266bcf6

In our netperf testing, TCP_STREAM56 shows about 20% or more performance regression on WSM/NHM and tigerton machines. The testing boot up both netserver and client on localhost. The testing command like this:
./snapshot_script_net TC_STREAM56 127.0.0.1

We found the following commit causes this issue. c377411f2494a931ff7facdbb3a6839b1266bcf6
Revert this commit will recover this regression on all machine.

Regards!
Alex