Ridiculously slow fat32 usb drive writes

relpats_eht · 06-23-2006, 09:48 PM

I recently noticed while writing files to my USB thumb-drive with a FAT32 file system that the write speed was averaging 8 KB/s, maxing at 32 KB/s and stalling frequently. I assumed the problem was the large amount of minuscule files was the problem, so I tried writing them all at once in an archived format, but that only raised the average to 12 KB/s and left the max alone. I can't think of a reason for the write speed to be this low, as all proper USB drivers are compiled into the kernel and my thumb-drives formatted with native Linux file systems perform fine, yet they all perform equally terrible formatted with FAT32. Thus, I ask, is this behavior normal (which I doubt), and if not, are there any suggestions which may improve this condition?

Tinkster · 06-23-2006, 10:52 PM

Hmmmm .. even if the thumb-drive is only USB 1.1 you'd expect around
150K/s ... do you get any odd messages in dmesg when doing those operations?

Cheers,
Tink

rkelsen · 06-24-2006, 07:42 AM

Quote:

Originally Posted by relpats_eht

all proper USB drivers are compiled into the kernel

Do your USB drives show up as /dev/sd* or /dev/ub* ?

relpats_eht · 06-30-2006, 07:03 PM

Sorry about the very late reply, I went on vaccation for a week and completely forgot about this.

To answer your questions, the thumbdrive is fullspeed and shows up as /dev/sd*. From what I can see, there is nothing odd in dmesg either.

Output of dmesg | grep usb:

Code:

usbcore: registered new driver usbfs
usbcore: registered new driver hub
usb usb1: configuration #1 chosen from 1 choice
usb usb2: configuration #1 chosen from 1 choice
usb usb3: configuration #1 chosen from 1 choice
usb usb4: configuration #1 chosen from 1 choice
usb usb5: configuration #1 chosen from 1 choice
usb 2-2: new low speed USB device using uhci_hcd and address 2
usb 2-2: configuration #1 chosen from 1 choice
usb 3-1: new full speed USB device using uhci_hcd and address 2
usb 3-1: configuration #1 chosen from 1 choice
usb 3-2: new low speed USB device using uhci_hcd and address 3
usb 3-2: configuration #1 chosen from 1 choice
usbcore: registered new driver cdc_acm
drivers/usb/class/cdc-acm.c: v0.25:USB Abstract Control Model driver for USB modems and ISDN adapters
drivers/usb/class/usblp.c: usblp0: USB Bidirectional printer dev 2 if 0 alt 0 proto 2 vid 0x03F0 pid 0x0904
usbcore: registered new driver usblp
drivers/usb/class/usblp.c: v0.13: USB Printer Device Class driver
usbcore: registered new driver usb-storage
input: USB HID v1.10 Joystick [Twin USB Joystick] on usb-0000:00:1d.0-2
input: USB HID v1.10 Mouse [Microsoft Microsoft 3-Button Mouse with IntelliEye(TM)] on usb-0000:00:1d.1-2
usbcore: registered new driver usbhid
drivers/usb/input/hid-core.c: v2.6:USB HID core driver
usbcore: registered new driver ov511
drivers/usb/media/ov511.c: v1.64 for Linux 2.5 : ov511 USB Camera Driver
drivers/usb/media/se401.c: SE401 usb camera driver version 0.24 registering
usbcore: registered new driver se401
usbcore: registered new driver mdc800
drivers/usb/image/mdc800.c: v0.7.5 (30/10/2000):USB Driver for Mustek MDC800 Digital Camera
usbcore: registered new driver microtekX6
usbcore: registered new driver usbled
usb 3-2: USB disconnect, address 3

AdaHacker · 06-30-2006, 08:12 PM

How are you mounting the drives? In particular, are you mounting them with the sync option?

For some reason (I don't know why) using synchronous writes on USB thumb drives results in miserable performance. I had the same problem. Simply removing the sync option from the drive's line in /etc/fstab, so that writes were done asynchronously, fixed the problem.

relpats_eht · 06-30-2006, 08:24 PM

Thank you. I was using sync. I have since removed it and hopefully the problem will be solved.