Skip to main content

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

18
  • Have you checked if there are any I/O errors or alike in the kernel log (dmesg / journalctl -k) when this occurs? Do all the other drives / partitions have different (type of) filesystems from the system one? (lsblk -f) Commented Oct 1 at 8:04
  • You may also want to iostat the corresponding drive to see if the speed you saw was really a "different speed". Commented Oct 1 at 8:06
  • @TomYan, I've updated my question to add my reply to your comments. It looks like the copied data is flushed from time to time at higher speeds, but I'm not sure. Anyway, even if it is so, are those 1.5 MB/s is OK for SSDs? Commented Oct 1 at 9:12
  • If you can temporarily spare one of your USB SSDs for testing, then try your tests on that drive while formatted as NTFS, vfat, and then ext4. Trim or secure-erase the SSD before partitioning and reformatting for each of those three filesystems. Compare the speeds for each fs - my bet is that ext4 will be fastest, vfat coming in a reasonably close second, and NTFS a distant third. NTFS support in Linux is not very good, and really shouldn't be used unless you're transferring files between Linux and Windows (and even then, vfat would be better unless you're dual-booting). Commented Oct 1 at 9:39
  • BTW, the USB type and speed (both ports and devices) also matters a lot. Using a USB 3 or USB 4 port with a USB drive capable of USB3/USB4 speeds will be a lot faster than either using the same drive in a USB2 port, or using an old USB2 drive in a more modern USB3/4 port. Also, as @TomYan asked, please add the ouput of lsblk -f (or blkid) to your question. Commented Oct 1 at 9:46