I used to be utilizing this information to attempt to carry out a big sequential write with FIO: https://chiadecentral.com/fio/
And modified the script on the backside of the web page as follows:
[global]
bs=1024K
iodepth=32
direct=1
ioengine=posixaio
group_reporting
time_based
title=seq
log_avg_msec=1000
bwavgtime=1000
filename=/Customers/*person*/Desktop
dimension=1000G
[wr_qd_256_128k_1w]
stonewall
bs=128k
iodepth=32
numjobs=1
rw=write
runtime=600
write_bw_log=seq_write_bw.log
It ran nicely, however I’m considerably confused as to how FIO works… and whether it is even significant working a benchmark on an inside system drive.
Principally – The check ran on the disk velocity of about 5.5GB/s for the primary 5 minutes, then started to decelerate, typical of an SSD. However I’m confused as to how FIO is doing the write check – because it doesn’t replenish the disk in any respect, what’s it writing to precisely and is it instantly eradicating any information that’s written?
The disk is just 1TB in dimension, so you’ll anticipate it to pretty quickly hit the ‘actual’ TLC state of the SSD, but it surely seemingly wrote for nearly 7 minutes at 5.5GB/s – nicely past the precise capability of the disk.
I ran the script anticipating FIO to jot down a big file sequentially – and I anticipated to see a decelerate within the disk loads prior to I did.
