Re: Google SoC idea

2005-06-10 Thread David Rhodus
On 6/8/05, Scott Long <[EMAIL PROTECTED]> wrote: > Ivan Voras wrote: > > Scott Long wrote: > > > >> Again, I'm not exactly sure how a generic mechanism can handle the > >> distinction of data vs. metadata vs. journal data. Also, what you > > > > > > I don't care about the distinction at this level

Re: Google SoC idea

2005-06-08 Thread Ivan Voras
Stephan Uphoff wrote: In my opinion the original proposal described a non volatile write cache using dedicated cache disks. Yes, I think that best describes what I have in mind. Here's the proposal that went to Google yesterday: http://geri.cc.fer.hr/~ivoras/proposal.pdf __

Re: Google SoC idea

2005-06-08 Thread Stephan Uphoff
On Tue, 2005-06-07 at 16:52, Scott Long wrote: > David Malone wrote: > > On Tue, Jun 07, 2005 at 09:40:05PM +0200, Pawel Jakub Dawidek wrote: > > > >>+> Does it make sense to do it this way? Is it worth applying for the SoC? > >> > >>Not sure. Basically this is simlar what softupdate does, I think

Re: Google SoC idea

2005-06-08 Thread Bruce R. Montague
Scott Long wrote: >Again, I'm not exactly sure how a generic mechanism can handle the >distinction of data vs. metadata vs. journal data. Also, what you Ivan Voras wrote: >I don't care about the distinction at this level - all data is treated >equal. Scott Long wrote: >But for j

Re: Google SoC idea

2005-06-08 Thread Scott Long
Ivan Voras wrote: Scott Long wrote: Again, I'm not exactly sure how a generic mechanism can handle the distinction of data vs. metadata vs. journal data. Also, what you I don't care about the distinction at this level - all data is treated equal. But for journalling to work, you must c

Re: Google SoC idea

2005-06-08 Thread Ivan Voras
Scott Long wrote: Again, I'm not exactly sure how a generic mechanism can handle the distinction of data vs. metadata vs. journal data. Also, what you I don't care about the distinction at this level - all data is treated equal :) ___ freebsd-hack

Re: Google SoC idea

2005-06-08 Thread Scott Long
Ivan Voras wrote: Scott Long wrote: An alternate SoC project that would be very useful is block-level snapshots. I'm not sure if I'll be able to retain the filesystem snapshot functionality in UFS with journalling enabled, so moving to doing the snapshots in the block layer would be a good w

Re: Google SoC idea

2005-06-08 Thread Scott Long
Eric Anderson wrote: Scott Long wrote: Richard Coleman wrote: Scott Long wrote: /me jumps up and down and waves his hands The problem with journalling at the block layer is that you pretty much become forced to journal metadata and data, since the block layer really doesn't know the dist

Re: Google SoC idea

2005-06-08 Thread Eric Anderson
Scott Long wrote: Richard Coleman wrote: Scott Long wrote: /me jumps up and down and waves his hands The problem with journalling at the block layer is that you pretty much become forced to journal metadata and data, since the block layer really doesn't know the distinction, and definitely

Re: Google SoC idea

2005-06-08 Thread Ivan Voras
Scott Long wrote: An alternate SoC project that would be very useful is block-level snapshots. I'm not sure if I'll be able to retain the filesystem snapshot functionality in UFS with journalling enabled, so moving to doing the snapshots in the block layer would be a good way to make up for

Re: Google SoC idea

2005-06-08 Thread Scott Long
Richard Coleman wrote: Scott Long wrote: /me jumps up and down and waves his hands The problem with journalling at the block layer is that you pretty much become forced to journal metadata and data, since the block layer really doesn't know the distinction, and definitely not in a filesyste

Re: Google SoC idea

2005-06-07 Thread Richard Coleman
Scott Long wrote: /me jumps up and down and waves his hands The problem with journalling at the block layer is that you pretty much become forced to journal metadata and data, since the block layer really doesn't know the distinction, and definitely not in a filesystem-independent way (yes, U

Re: Google SoC idea

2005-06-07 Thread David Malone
> The problem with journalling at the block layer is that you pretty much > become forced to journal metadata and data, since the block layer really > doesn't know the distinction, Definitely - I guess I should have stated that explicitly. > Full journalling has many drawbacks from the viewpoin

Re: Google SoC idea

2005-06-07 Thread Scott Long
David Malone wrote: On Tue, Jun 07, 2005 at 09:40:05PM +0200, Pawel Jakub Dawidek wrote: +> Does it make sense to do it this way? Is it worth applying for the SoC? Not sure. Basically this is simlar what softupdate does, I think. From another point of view softupdates are only available for UF

Re: Google SoC idea

2005-06-07 Thread David Malone
On Tue, Jun 07, 2005 at 09:40:05PM +0200, Pawel Jakub Dawidek wrote: > +> Does it make sense to do it this way? Is it worth applying for the SoC? > > Not sure. Basically this is simlar what softupdate does, I think. > From another point of view softupdates are only available for UFS. > You probabl

Re: Google SoC idea

2005-06-07 Thread Pawel Jakub Dawidek
On Mon, Jun 06, 2005 at 06:11:23PM +0200, Ivan Voras wrote: +> I have an idea that I could implement through Google's "Summer of Code" +> project, but as I have little experience with stuff it involves (kernel +> programming / disks / filesystem optimization), so I expect any answer +> from "It

Google SoC idea

2005-06-06 Thread Ivan Voras
I have an idea that I could implement through Google's "Summer of Code" project, but as I have little experience with stuff it involves (kernel programming / disks / filesystem optimization), so I expect any answer from "It won't work" or "It's useless" to "It can't be done". :) The idea is t