On Monday 17 February 2025 23:12:59 Greenwich Mean Time Frank Steinmetzger wrote: > Am Sun, Feb 16, 2025 at 01:41:10PM +0000 schrieb Michael:
> > I just formatted a USB 2.0 8GB stick with mke2fs as a test. It took 46 > > seconds. Extrapolating for your 64GB partition it should take ~ 6 > > minutes. A full 256GB drive would take 24.5 minutes. I expect a drive > > enjoying USB 3.0 transfer speeds to take way less than this. > > Your expectations have some vague assumptions: Yes, I was just trying to point loosely at the order of magnitude as an indication of something being faulty in the USB Philip was trying to format. > - the stick achieves full speed of its USB protocol I haven't yet found a USB stick which reaches the maximum USB protocol speeds. The old USB 2.0 stick I used will not go above 11-12MBps during large writes and around twice as fast on sustained reads, well below the advertised USB 2.0 speed of 60MBps. > - it keeps that throughput over time Once buffers are saturated write operations will settle at what the device can achieve. I have observed writes slow down when the empty space left is getting low. > - the data written grows linearly with the FS size Usually larger USB drives of the same brand/model tend to have slightly better speeds - at least this is what the data sheet for the USB drive Philip bought claims. > USB sticks often have a bad thermal design and throttle down once they get > hot. And hot they get, especially the high-speed ones. > > If you are curious, use a program like dool or glances to observe the > device’s actual write throughput during formatting. htop can show this, too, > but you first need to add a display meter in its config. I use gkrellms on GUI, and 'watch -d vmstat -d -S M' when on a console, but I've also used htop.
signature.asc
Description: This is a digitally signed message part.
