Comment 20 for bug 709245
Revision history for this message
| Michael Sparmann (theseven) wrote Re: panda: USB disk IO slow | #20 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
a60fb26
(Get the code!)
When running just dd at 1MB block size, I get ~5.5MB/s.
When running with sudo ping -q -i.0100 localhost in the background, ping causes 1% CPU load, and I get ~10MB/s at 1MB block size and these ping statistics:
6228 packets transmitted, 6227 received, 0% packet loss, time 97329ms
rtt min/avg/max/mdev = 0.000/0.
When running with sudo ping -q -i.0099 localhost in the background, ping causes 100% (one core) CPU load, and I get ~20MB/s at 1MB block size and these ping statistics:
10339 packets transmitted, 10337 received, 0% packet loss, time 93032ms
rtt min/avg/max/mdev = 0.000/0.
Now why does ping process about twice as many packets but use 100 times more CPU (and get better response times) when I reduce the delay by just one percent? There's something fishy going on here...