0

I've got the following problem with my old WinXP machine (1.66 GHz Intel Core 2 Duo, 2 GB RAM, 500 GB HDD):

After a wakeup from hibernation, the HD goes crazy for at least a minute. It's especially bad if the computer was hibernated for a long time. The guilty processes are "System" with a Page Fault Delta of at least 3000 and constant I/O Delta at about 100 and "csrss.exe" with similar values. (I know this doesn't sound too impressive; the highest PF Delta I've seen is about 80.000 after a fresh boot, but it's enough to stop practically everything else).

Is this really necessary? Can I remedy the condition?

Improve this question

1 Answer 1

Reset to default
2

When you put a windows machine into hibernation, you dump the contents of memory to disk. When you come out of hibernation, windows will page fault on every memory read, since it has load that page from disk. If the memory gets dumped to non-sequential tracks on the disk or something, it's going to have to seek all over the place while state's restored. This sounsd pretty normal to me.

Improve this answer
2
  • MrTuttle is completely right. Your only option is to use standby instead of hibernate: in standby, power is still applied so RAM retains its contents. Also, the time hibernated shouldn't matter too much: you may be looking at a case of expected-results-bias there.
    – Michael Lowman
    Commented Jul 7, 2011 at 20:04
  • But I thought if the whole memory was written on disk upon hibernation and restored afterwards, I shouldn't have more PFs than before... or is the memory not copied/restored 1:1?
    – Hinton
    Commented Jul 9, 2011 at 18:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .