| 1 | # Sample job file demonstrating how to use zoned random distributionss |
| 2 | # to have skewed random accesses. This example has 50% of the accesses |
| 3 | # to the first 5% of the file (50/5), 30% to the next 15% (30/15), and |
| 4 | # finally 20% of the IO will end up in the remaining 80%. |
| 5 | [zones] |
| 6 | size=2g |
| 7 | direct=1 |
| 8 | bs=4k |
| 9 | rw=randread |
| 10 | norandommap |
| 11 | random_distribution=zoned:50/5:30/15:20/ |
| 12 | |
| 13 | # It's also possible to use zoned_abs to specify absolute sizes. For |
| 14 | # instance, if you do: |
| 15 | # |
| 16 | # random_distribution=zoned_abs:50/10G:30/100G:20/500G |
| 17 | # |
| 18 | # Then 50% of the access will be to the first 10G of the drive, 30% |
| 19 | # will be to the next 100G, and 20% will be to the next 500G. |
| 20 | |
| 21 | # The above applies to all of reads/writes/trims. If we wanted to do |
| 22 | # something differently for writes, let's say 50% for the first 10% |
| 23 | # and 50% for the remaining 90%, we could do it by adding a new section |
| 24 | # after a a comma. |
| 25 | |
| 26 | # random_distribution=zoned:50/5:30/15:20/,50/10:50/90 |