Hello List, We have a network appliance that is currently based on 4.9. We are in the process of releasing a new version based on 6.1 stable. In our testing using nttcp thru the appliance we see insignifant difference in thruput between the 2 versions in a controlled environment - aproximately 94mbs on a 100mb lan. We have a person that is testing the both system inhouse surfing out over the internet on our T1 link and he complains that he is consistently seeing the 6.1 version being much slower than the 4.9 version (on the same hardware). He has been comparing the 6.1 system to 4.9 system for a couple of weeks and continues to insist the 6.1 version is much slower. Are there any sysctl tunables that may affect performance going over the internet with a slower link, dropped packets, etc that could cause this? Any ideas would be appreciated. Steve -- "They that give up essential liberty to obtain temporary safety, deserve neither liberty nor safety." (Ben Franklin) "The course of history shows that as a government grows, liberty decreases." (Thomas Jefferson)
> We have a network appliance that is currently based on 4.9. We are in > the process of releasing > a new version based on 6.1 stable. > > In our testing using nttcp thru the appliance we see insignifant > difference in thruput between the 2 > versions in a controlled environment - aproximately 94mbs on a 100mb lan. > > Are there any sysctl tunables that may affect performance going over the > internet > with a slower link, dropped packets, etc that could cause this?Have you tried options DEVICE_POLLING in the kernel? If you want to try yuo can find a section in the file NOTES located the same place as your kernel. Is the kernel cusomized? Have you enabled -O2 as optimization? Do you consider release 6.2 rather than 6.1? -- regards Claus When lenity and cruelty play for a kingdom, the gentlest gamester is the soonest winner. Shakespeare
Stephen Clark wrote:> Hello List, > > We have a network appliance that is currently based on 4.9. We are in > the process of releasing > a new version based on 6.1 stable.You are going to get asked this, so I'll ask first. Whats the reason behind not running a more recent STABLE? I understand developing a product on a moving platform is not ideal, but its going to be mentioned!> In our testing using nttcp thru the appliance we see insignifant > difference in thruput between the 2 > versions in a controlled environment - aproximately 94mbs on a 100mb lan. > > We have a person that is testing the both system inhouse surfing out > over the internet on our T1 > link and he complains that he is consistently seeing the 6.1 version > being much slower than the > 4.9 version (on the same hardware). > He has been comparing the 6.1 system to 4.9 system for a couple of weeks > and continues to insist the 6.1 version is much slower.You don't mention what the appliance actually does beyond just moving packets about? Surfing implies some sort of proxy or gateway device?> Are there any sysctl tunables that may affect performance going over the > internet > with a slower link, dropped packets, etc that could cause this? > > Any ideas would be appreciated. >Dominic
Why are you releasing a new product on an already unsupported version? 6.2 is the version you really need to be moving to not 6.1. With respect to your actual question you provide no real details so there's no real answers. You need to provider info on hardware, configuration and application + concrete metrics otherwise your going to get no where. That said I'd suggest as a total and utter guess setting: net.inet.tcp.inflight.enable=0 Regards Steve ----- Original Message ----- From: "Stephen Clark" <Stephen.Clark@seclark.us> To: <freebsd-stable@freebsd.org> Sent: Friday, May 25, 2007 1:05 PM Subject: network performance 6.1 stable vs 4.9> Hello List, > > We have a network appliance that is currently based on 4.9. We are in > the process of releasing > a new version based on 6.1 stable. > > In our testing using nttcp thru the appliance we see insignifant > difference in thruput between the 2 > versions in a controlled environment - aproximately 94mbs on a 100mb lan. > > We have a person that is testing the both system inhouse surfing out > over the internet on our T1 > link and he complains that he is consistently seeing the 6.1 version > being much slower than the > 4.9 version (on the same hardware). > He has been comparing the 6.1 system to 4.9 system for a couple of weeks > and continues to insist the 6.1 version is much slower. > > Are there any sysctl tunables that may affect performance going over the > internet > with a slower link, dropped packets, etc that could cause this? > > Any ideas would be appreciated.===============================================This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmaster@multiplay.co.uk.
On Friday 25 May 2007 15:05, Stephen Clark wrote:> We have a person that is testing the both system inhouse surfing out > over the internet on our T1 link and he complains that he is consistentlyseeing the 6.1 version being much slower than the> 4.9 version (on the same hardware).FreeBSD cannot handle 1.5Mbps? OK it's actually 3Mbps. Doesn't this sound a bit weird(or funny)? Can you please elaborate? Nikos
Stephen Clark wrote:> Hello List, > > We have a network appliance that is currently based on 4.9. We are in > the process of releasing > a new version based on 6.1 stable. > > In our testing using nttcp thru the appliance we see insignifant > difference in thruput between the 2 > versions in a controlled environment - aproximately 94mbs on a 100mb lan. > > We have a person that is testing the both system inhouse surfing out > over the internet on our T1 > link and he complains that he is consistently seeing the 6.1 version > being much slower than the > 4.9 version (on the same hardware). > He has been comparing the 6.1 system to 4.9 system for a couple of weeks > and continues to insist the 6.1 version is much slower. > > Are there any sysctl tunables that may affect performance going over > the internet > with a slower link, dropped packets, etc that could cause this? >Well, oddly enough I've playing with 6.1's perf as it relates to gigE lans so let me pass on a few things * polling only really helps if your nic is generating lot's of interrupts, or is having to compete with something that does. In fact, setting "polling" on an interface could make things seem slightly LESS responsive for small requests * nic chipset selection is important, but probably not for a f/w dealing with t1/broadband speeds * Don't even get down tweaking tcp send and recv buffers. You have no idea what the BDP will be on your WAN link. Same thing for jumbo frames on the inside link. o Having said that, This is what is in my sysctl.conf file. It does matter in a gigE lan, but probably not for a SMB firewall thats only got a t1 on the WAN side. kern.ipc.maxsockbuf=8192000 net.inet.tcp.sendspace=262144 net.inet.tcp.recvspace=262144 * make sure you set net.inet.tcp.rfc1323: 1 (most likely the default) * play with net.inet.tcp.inflight.enable (0 or 1), it never made a diff in my gigE lan testing * get real data using iperf (in ports/benchmarking) if you go to the iperf website, they have binaries for windows o XP's default network tuning is beyond bad, but your customers most likely aren't tweaking their registries either * use netstat -m to look at your buffer usage, particularly if you're dropping packets * depending on how much memory you have you might want to jump up kern.ipc.nmbclusters, but only if you seem to be dropping packets. * verify that your nics are setting speed and duplex correctly * which firewall package are you using? * use ethereal/wireshark to examine your net flow. Alot of tcp resets and retransmits can make a big impact if TCP is constantly having to resync. jim
On Fri, May 25, 2007 at 08:05:15AM -0400, Stephen Clark wrote:> Hello List, > > We have a network appliance that is currently based on 4.9. We are in > the process of releasing > a new version based on 6.1 stable. > > In our testing using nttcp thru the appliance we see insignifant > difference in thruput between the 2 > versions in a controlled environment - aproximately 94mbs on a 100mb lan. > > We have a person that is testing the both system inhouse surfing out > over the internet on our T1 > link and he complains that he is consistently seeing the 6.1 version > being much slower than the > 4.9 version (on the same hardware). > He has been comparing the 6.1 system to 4.9 system for a couple of weeks > and continues to insist the 6.1 version is much slower. > > Are there any sysctl tunables that may affect performance going over the > internet > with a slower link, dropped packets, etc that could cause this? > > Any ideas would be appreciated. > > SteveI do not believe that I have seen any description of the hardware involved. That could play a factor.
On Fri, 25 May 2007, Stephen Clark wrote:> We have a network appliance that is currently based on 4.9. We are in the > process of releasing a new version based on 6.1 stable. > > In our testing using nttcp thru the appliance we see insignifant difference > in thruput between the 2 versions in a controlled environment - aproximately > 94mbs on a 100mb lan. > > We have a person that is testing the both system inhouse surfing out over > the internet on our T1 link and he complains that he is consistently seeing > the 6.1 version being much slower than the 4.9 version (on the same > hardware). He has been comparing the 6.1 system to 4.9 system for a couple > of weeks and continues to insist the 6.1 version is much slower. > > Are there any sysctl tunables that may affect performance going over the > internet with a slower link, dropped packets, etc that could cause this? > > Any ideas would be appreciated.Steve, The first thing I'd do is try a double-blind test for your testers -- don't tell them which version is running, and then compare performance complaints with/without. This would let you know if there's actually a difference. The main piece of advice I give people when working with 6.x is to consider turning on net.isr.direct, which enables direct dispatch in the network stack. With 4.x, I get lower forwarding and processing latency than 6.x unless I enable this. However, my recollection is that you don't want to turn it on on releases before 6.1, and I would really be most comfortable turning it on with 6.2 and later. In FreeBSD 7.0, net.isr.direct is the default. You might give that a try and see if it has an effect, but I'd see about getting some sort of objective testing of performance going to confirm that this isn't a subjectivity issue. Robert N M Watson Computer Laboratory University of Cambridge> > Steve > > -- > > "They that give up essential liberty to obtain temporary safety, deserve > neither liberty nor safety." (Ben Franklin) > > "The course of history shows that as a government grows, liberty decreases." > (Thomas Jefferson) > > > > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" >