I've disabled hibernation, and removed the sleep image file. I shutdown my Macbook Pro (15" retina), unticked the box to open windows on restart, and started my MacBook again. As soon as I'd logged on I went to Activity Monitor and saw kernal_task is using 1GB of RAM.

This is a problem as I use parallels and when using Parallels, Parallels itself uses ~3.5GB of RAM and the kernel_task jumps up to over 2GB of RAM. So I have hardly any RAM left for the rest of the computer and it slows down to a crawl.

This has only been a problem since recently. Before I could run Parallels, Mail, Tweetbot, and Safari all at once (and in Parallels on Windows 7 I was using Visual Studio 2012 and/or SimCity). It was only once I ran SimCity along with the rest that it started slowing down. But now it slows down as I say above with nothing open in Windows 7 in Parallels, let alone VS2012 or SimCity.

Is there a way to see what the heck kernel_task is doing and to cut it down?

Will reinstalling OSX work (without wiping my hard drive first so I can keep my files)? I'd rather not do this as it takes about 24 hours to download OSX Lion on my connection.

EDIT: It's now ridiculous I can't open Parallels at all because after a few minutes the computer runs out of RAM and I'm forced to crash it by holding down the power button. I will offer a 500+ bounty (Can't atm because of the stupid 2 day wait) if that helps.

  • similar issue here: when running a memory-instensive process, kernel_task consumes about 2X of process' memory causing the whole system to freeze.
    – lynxoid
    May 15, 2014 at 18:30

1 Answer 1


I know your post is from a while back, but if this is still an issue, try my suggestions in this post. If those don't work, D.W. posted some other useful links in his comments.

Hope you get this worked out :)


  • 1
    Please provide a description as to what exactly your method does.
    – killswitch
    Aug 3, 2014 at 6:39
  • Basically I had three suggestions there. The first one reduces the amount of files for OSX to cache into RAM and consequently lowers the ram usage. This will probably be most helpful fix for ram related issues, but the second one is pretty good too. The second suggestion has two parts (only the first one will, probably, be relevant here) this one shows a way to automatically detect and fix pc issues and, in my experience, helps improve performance quite a lot. The third one is just a simple diagnostic tool that can help us narrow down where the problem could be coming from. :)
    – Vladimir
    Aug 3, 2014 at 7:03
  • Actually now that one of my answers was upvoted, the order on the page is not the same as when I wrote the above comment. Now 1 and 2 have been switched places. :)
    – Vladimir
    Aug 3, 2014 at 7:24

You must log in to answer this question.

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