[colug-432] 200 line kernel patch

[Jeff Frontz]

So I spent a few minutes trying to figure out what they were trying to do
with this -- is it just that they're putting the "interactive" processes in
their own scheduling group and this (or all?) scheduling groups get
scheduled prior to hoi polloi ?

How is that any different from making the login shell be RT (and a suitably
high priority)?

More importantly, what does this do to the actual through-put of the system?
 Doesn't it cause all sorts of crazy starvation that just ends up making the
overall system sluggish (because of all sorts of retries and timeouts that
have to get handled)?

Jeff


On Mon, Dec 13, 2010 at 9:44 PM, Bill Baker <bill_chris at earthlink.net>wrote:

> Getting back to the subject of Linux...
>
> Has anyone tried out the new 200 line kernel patch that supposedly
> improves responsiveness?  I just added the alternative script to
> my /etc/rc.local file that's supposed to do the same thing detailed
> here:
> http://www.webupd8.org/2010/11/alternative-to-200-lines-kernel-patch.html
>
> So far I don't see much of an improvement, but I'll continue to use it
> to see if I do notice anything.  I was just checking to see if anyone
> had done the actual patch and if there was any dramatic improvement.
>
> _______________________________________________
> colug-432 mailing list
> colug-432 at colug.net
> http://lists.colug.net/mailman/listinfo/colug-432
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.colug.net/pipermail/colug-432/attachments/20101215/86d6e650/attachment.html