Re: [HACKERS] hint bit cache v5

2011-05-11 Thread Merlin Moncure
On Wed, May 11, 2011 at 12:40 PM, Simon Riggs wrote: > On Wed, May 11, 2011 at 4:38 PM, Merlin Moncure wrote: > >> Following are results that are fairly typical of the benefits you >> might see when the optimization kicks in.  The attached benchmark just > >> [hbcache] >> real    3m35.549s > >> [

Re: [HACKERS] hint bit cache v5

2011-05-11 Thread Simon Riggs
On Wed, May 11, 2011 at 4:38 PM, Merlin Moncure wrote: > Following are results that are fairly typical of the benefits you > might see when the optimization kicks in.  The attached benchmark just > [hbcache] > real    3m35.549s > [HEAD] > real    4m24.216s These numbers look very good. Thanks

Re: [HACKERS] hint bit cache v5

2011-05-11 Thread Merlin Moncure
On Wed, May 11, 2011 at 10:38 AM, Merlin Moncure wrote: > One thing I need to test is how much benefit you'll see with wider records. The results are a bit better, around 25% using a similar methodology on ~ 1k wide records. > I think I'm gonna revert the change to cache invalid bits. I just > d

Re: [HACKERS] hint bit cache v5

2011-05-11 Thread Merlin Moncure
On Tue, May 10, 2011 at 11:59 AM, Simon Riggs wrote: > On Mon, May 9, 2011 at 5:12 PM, Merlin Moncure wrote: > >> I'd like to know if this is a strategy that merits further work...If >> anybody has time/interest that is.  It's getting close to the point >> where I can just post it to the commit f

Re: [HACKERS] hint bit cache v5

2011-05-10 Thread Merlin Moncure
On Tue, May 10, 2011 at 11:59 AM, Simon Riggs wrote: > On Mon, May 9, 2011 at 5:12 PM, Merlin Moncure wrote: > >> I'd like to know if this is a strategy that merits further work...If >> anybody has time/interest that is.  It's getting close to the point >> where I can just post it to the commit f

Re: [HACKERS] hint bit cache v5

2011-05-10 Thread Simon Riggs
On Mon, May 9, 2011 at 5:12 PM, Merlin Moncure wrote: > I'd like to know if this is a strategy that merits further work...If > anybody has time/interest that is.  It's getting close to the point > where I can just post it to the commit fest for review.  In > particular, I'm concerned if Tom's ear

[HACKERS] hint bit cache v5

2011-05-09 Thread Merlin Moncure
Attached is an updated version of the 'hint bit cache'. What's changed: *) 'bucket' was changed to 'page' everywhere *) rollup array is now gets added during 'set', not the 'get' (pretty dumb the way it was before -- wasn't really dealing with non-commit bits yet) *) more source comments, includin