curl is a useful tool if you’re working with HTTP. I’m fond of the -w option: it prints all kind of information about the transfert, including timing:

$ curl -s -o /dev/null -w '
url_effective: %{url_effective}
http_code: %{http_code}
time_total: %{time_total}
time_namelookup: %{time_namelookup}
time_connect: %{time_connect}
time_pretransfer: %{time_pretransfer}
time_starttransfer: %{time_starttransfer}
size_download: %{size_download}
size_upload: %{size_upload}
size_header: %{size_header}
size_request: %{size_request}
speed_download: %{speed_download}
speed_upload: %{speed_upload}
content_type: %{content_type}'

http_code: 301
time_total: 0.062
time_namelookup: 0.038
time_connect: 0.045
time_pretransfer: 0.045
time_starttransfer: 0.062
size_download: 218
size_upload: 0
size_header: 320
size_request: 153
speed_download: 3504.000
speed_upload: 0.000