System Unresponsive During Long File Transfers and Processes

That's the thing that's odd  -- the system performs phenominally when not conducting large file transfers and so forth, with apps and windows opening and closing instantaneously.  It's just when I'm trying to conduct normal browsing or terminal use while also performing these larger operations that I run into a problem.  I'll keep an eye out for similar behaviors and report my findings.

Now that you mention it, fabsh, the file transfers to/from my external HD were via Thunar -- and the .tar file concatenation/unpacking processes, which I reported as going much more smoothly, were done in the terminal.  Could Thunar be the culprit?  Maybe I'll install nautilus and test the same actions.

Macmeister--

Thanks so much for providing the feedback.  If you're willing and/or daring enough to backup your system and load the new 8.10.01 #! as your primary OS, it might provide better insight into this idiosyncrasies of this particular version of the distro.  This seems like a select enough issue that it may be version- rather than distro-specific.

If you'd like a nice little script to help automate your backup process please let me know -- I've written a little script called backmeup that speeds up the backup process.

fabsh -- I'm also fulltime on #! -- the whole thing has been an unexpected renaissance for me.

Mmmm, sounds like interrupt problems. Maybe it's kernel specific. That would explain how Mint is OK but #! is slow.

Like my machine,  Firewire and my Nvidia card on the same interrupt, so when i'm doing heavy firewire transfers I get all sorts of skipping and jumping. Put the output of /proc/interrupts onto pastebin so we can have a gander

[edit]

Also, check /proc/cpuinfo to see if #! is using both cores

Ouch, you've got usb, audio, and graphics on one interrupt.

But yes, your right, it should balance more between the two cores. Check this out. may give you some handy hints to improve matters.

[edit]

Oh yes, looks like Ubuntu kernels have this issue by default.

CONFIG_ACPI_HOTPLUG_CPU=y

hey guys, im getting basically the same as Kestrel, all the load is on one cpu

~$ more /proc/interrupts
           CPU0       CPU1       
  0:     147690          0   IO-APIC-edge      timer
  8:         70          0   IO-APIC-edge      rtc0
  9:       6709          0   IO-APIC-fasteoi   acpi
 14:       7133          0   IO-APIC-edge      ata_piix
 15:          0          0   IO-APIC-edge      ata_piix
 16:        190          0   IO-APIC-fasteoi   uhci_hcd:usb4, fglrx[0]@PCI:1:0:0
 17:      64175          0   IO-APIC-fasteoi   ath
 18:         47          0   IO-APIC-fasteoi   uhci_hcd:usb3
 19:      15899          0   IO-APIC-fasteoi   uhci_hcd:usb2, ata_piix, ohci1394
 22:        429          0   IO-APIC-fasteoi   HDA Intel
 23:      47522          0   IO-APIC-fasteoi   uhci_hcd:usb1, ehci_hcd:usb5
219:       9959          0   PCI-MSI-edge      eth0
NMI:          0          0   Non-maskable interrupts
LOC:      33795     104398   Local timer interrupts
RES:      26422      46611   Rescheduling interrupts
CAL:      13953       6970   function call interrupts
TLB:       6430       2507   TLB shootdowns
SPU:          0          0   Spurious interrupts
ERR:          0
MIS:          0

I took a look at the link provided, and it looks like I have to install a new kernel?  how in the world do i do that without killing my #! install? or is there some easier solution?