Randy, On 2020/01/15 3:25, Randy Dunlap wrote: > Hi Damien, > > Here are a few editorial comments for you... Thanks ! All fixed. [...] >> +For sequential write zone files, the file size changes as data is appended at >> +the end of the file, similarly to any regular file system. >> + >> +# dd if=/dev/zero of=/mnt/seq/0 bs=4096 count=1 conv=notrunc oflag=direct >> +1+0 records in >> +1+0 records out >> +4096 bytes (4.1 kB, 4.0 KiB) copied, 1.05112 s, 3.9 kB/s > > Still slow. You don't want to change that? Good catch. I thought I had fixed that. Here is the updated dd run, after making sure that the disk has woken up from low power state before running: dd if=/dev/zero of=/mnt/seq/0 bs=4096 count=1 conv=notrunc oflag=direct 1+0 records in 1+0 records out 4096 bytes (4.1 kB, 4.0 KiB) copied, 0.00044121 s, 9.3 MB/s The HDD write cache is on and empty at the time of running this, which explains the much lower I/O time. -- Damien Le Moal Western Digital Research
