Block devices using a non-optimal block size and causing excessive CPU overhead

Issue

• Device mapper shows an average request size of 512 bytes (1 sector) and an average queue size of ~800,000 for background cache I/O.

Device:         rrqm/s   wrqm/s     r/s     w/s    rkB/s    wkB/s avgrq-sz avgqu-sz   await  svctm  %util
...
dm-0              0.00     0.00    0.00 379732.00     0.00 189866.00     1.00 742618.96  349.45   0.00 100.00
dm-0              0.00     0.00    0.00 24685.00     0.00 12342.50     1.00 867568.40 11955.14   0.04 100.00
dm-0              0.00     0.00    0.00 70797.00     0.00 35398.50     1.00 803960.57 4986.36   0.01 100.00
dm-0              0.00     0.00    0.00 74472.73     0.00 37236.36     1.00 780756.23 5746.69   0.01 101.01
dm-0              0.00     0.00    0.00 72998.02     0.00 36499.01     1.00 736369.45 6792.40   0.01  99.01
dm-0              0.00     0.00    0.00 74472.73     0.00 37236.36     1.00 723803.84 7759.93   0.01 101.01
dm-0              0.00     0.00    0.00 246027.72     0.00 123013.86     1.00 809971.51 2115.03   0.00  99.01
dm-0              0.00     0.00    0.00    0.00     0.00     0.00     0.00 816929.14    0.00   0.00 100.00
...

This is far from optimal and is causing excessing CPU overhead

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
...
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
           0.00    0.00  100.00    0.00    0.00    0.00
...