The system load is ~1.15, and mysqld is highest on the list of processes but is only 1-2%, which I assume means that the cpu is being blocked somewhere due to IO (?).
Disk write performance inside the VM where Bacula and mysql runs is around 90MB/second, which is more than sufficient to insert a few hundred thousand rows.
Digging further... I had make the assumption that low kb/second of disk activity meant that the disk subsystem was idle. Russell pointed out that %util when iostat is called with -x and it shows high io load. In dom0 when mostly idle: # dd if=/dev/zero of=test.bin bs=1024 count=10240 oflag=direct 10240+0 records in 10240+0 records out 10485760 bytes (10 MB) copied, 2.00537 s, 5.2 MB/s In domu when idle: # dd if=/dev/zero of=test.bin bs=1k count=1024 oflag=direct 1024+0 records in 1024+0 records out 1048576 bytes (1.0 MB) copied, 0.262493 s, 4.0 MB/s In domu when slow mysql query is running: # dd if=/dev/zero of=test.bin bs=1k count=1024 oflag=direct 1024+0 records in 1024+0 records out 1048576 bytes (1.0 MB) copied, 20.6291 s, 50.8 kB/s Which does indicate that the io is struggling when the mysql query is running... I'm converting to xfs from ext3 to see if there is something gone bad with the filesystem. James