There’s a viewpoint issue with these measurements. ATS will record when the
last byte was written to the inbound socket. This will always appear longer
from the client point of view because those bytes will take time to move from
the kernel socket buffers to the client and this depends strongly on the nature
of the network path from ATS to the client. Generally kernel socket buffers are
about 1M to 2M, therefore if the content is less than that ATS will write it
out effectively instantaneously because it fits in the buffer. However if
there’s a slow link between ATS and the client it would easily be a few hundred
milliseconds for the data to transit that link.
HTTP Timing breakdown:
Connection setup: 1.85
Request sent: 30.88
Wait response: 49.44
Content Download: 423.36
TOTAL: 505.53 ms
I cannot find any correlation.
Could you tell if it is possible to achieve this?
Thanks in advance.
Nuno Baeta
[email protected]<mailto:[email protected]>
