Stephen Wille Padnos wrote:
> Jon Elson wrote:
>   
>
> That's true, but is also generally not needed.  Since you don't care how 
> long it takes to get from switch 1 to switch 2 (unless you're 
> considering eliminating one or more of them), you don't need to see the 
> intermediate times.  The time ping reports is the cumulative time.  
> There's a program called "mtr" which gives a very nice display of ping 
> times, by the way.
>   
Well, I get huge differences between ping and traceroute for remote 
nodes.  The ping
might be 68 ms, but when I do traceroute, I get several nodes showing 
100+ ms delay
EACH.  Now, maybe it is traceroute that is giving the horribly inflated 
numbers.
>>    Still, 300 uS is not such a great time if you need 3
>> messages to propagate
>> within one millisecond.
>>    
>>     
> Note that ping is one outbound and one reply packet.  The pinging 
> machine has to get a response back before it can know the delay time.  I 
> don't know how ping calculates the delay time though - start of send to 
> end of receive, end of send to end of receive, compensation for the 
> number of bytes, etc.
>   
I would assume it times from when the packet is queued to when the 
checked reply packet
is handed back to ping.  But, this is very apropos to a real time 
controller sending a
request to a device and waiting to get a response back.  (Ask for 
encoder position,
get the position back, then send velocity update.)

Jon

------------------------------------------------------------------------------
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to