I had dedicated LAN cards with crossover cables!
Ping times were low, so it didn't make sense to me. I must have buggered something up.
May-12-2007 16:12
David Adams
Supreme tranceaddict
Registered: Jun 2006
Location: Texas, USA
A ping response determines latency at the network layer, not the application layer. The latency between the the FX-Teleport server and client may be entirely different. There is much more data that has to flow between the FX-Teleport client and server than that which has to flow between a simple ping.
I just wanted to set that record straight. Regardless, I don't see much latency at all with FX-Teleport across my 100Mbps LAN network - a little less than 10 msec.
wonder if any advantage is gained by using teaming and port trunking on a gig switch =)
Also, I bet if you did some MTU + TCP/IP stack tuning you'd squeeze out even better performance out of it
___________________
quote:
No one wants to die. Even people who want to go to heaven don’t want to die to get there. And yet death is the destination we all share. No one has ever escaped it. And that is as it should be, because Death is very likely the single best invention of Life. It is Life’s change agent. It clears out the old to make way for the new. Right now the new is you, but someday not too long from now, you will gradually become the old and be cleared away. Sorry to be so dramatic, but it is quite true.
--Steve Jobs (1955 - 2011)
May-16-2007 20:08
David Adams
Supreme tranceaddict
Registered: Jun 2006
Location: Texas, USA
quote:
Originally posted by emc^2
wonder if any advantage is gained by using teaming and port trunking on a gig switch =)
Also, I bet if you did some MTU + TCP/IP stack tuning you'd squeeze out even better performance out of it
The performance gain would probably be negligible. It is already indistinguishable from a locally hosted VST.
What would port trunking do? That just takes different Vlans and puts them on the same physical port i.e. dot1q or isl (Cisco proprietary) trunking. Do you mean etherchannel?
Over a LAN, I'm not sure what MTU modification would do for you since the response time between end nodes is already low (usually <1 msec). MTU modification can make the world of difference over high speed, high latency circuits like big WAN pipes e.g. DS3/E3 or OC3/STM-1, etc. When I say high latency, I mean as compared to what is seen on a LAN. WAN communications can usually always benefit from some sort of MTU modification. I'm a network engineer for a large company and have had several instances where customers are paying for OC-3 WAN circuits, but a single FTP can only squeeze about 5 Mbps out of it. Once I tell them about MTU modification, their FTP transfer speeds can go way up. There's actually a formula out there - perhaps I can still find it if anyone is interested.
