Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-12 Thread Aneesh Kumar K.V
Namjae Jeon writes: > From: Namjae Jeon > > This patch is based on suggestion by Wu Fengguang: > https://lkml.org/lkml/2011/8/19/19 > > kernel has mechanism to do writeback as per dirty_ratio and dirty_background > ratio. It also maintains per task dirty rate limit to keep balance of > dirty pag

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-11 Thread Fengguang Wu
> I will update simple table for this result in changelog in patch. That would be good, thanks a lot for the testing! Thanks, Fengguang -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://v

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-11 Thread Namjae Jeon
2012/9/12, Namjae Jeon : > 2012/9/12, Fengguang Wu : >> On Wed, Sep 12, 2012 at 08:12:40AM +0900, Namjae Jeon wrote: >>> >> >>> >> To be frank, no realistic NFS servers will use USB disk as backing >>> >> storage. So that rational for reducing "initial" delays is weak. >>> >> Continuous write perfo

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-11 Thread Namjae Jeon
2012/9/12, Fengguang Wu : > On Wed, Sep 12, 2012 at 08:12:40AM +0900, Namjae Jeon wrote: >> >> >> >> To be frank, no realistic NFS servers will use USB disk as backing >> >> storage. So that rational for reducing "initial" delays is weak. >> >> Continuous write performance to HDD is much more impor

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-11 Thread Fengguang Wu
On Wed, Sep 12, 2012 at 08:12:40AM +0900, Namjae Jeon wrote: > >> > >> To be frank, no realistic NFS servers will use USB disk as backing > >> storage. So that rational for reducing "initial" delays is weak. > >> Continuous write performance to HDD is much more important. Do you > >> have numbers f

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-11 Thread Namjae Jeon
>> >> To be frank, no realistic NFS servers will use USB disk as backing >> storage. So that rational for reducing "initial" delays is weak. >> Continuous write performance to HDD is much more important. Do you >> have numbers for that? > > Actually, we use USB HDD and USB Flash devices at NFS serv

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-10 Thread Namjae Jeon
2012/9/11, Fengguang Wu : > Hi Namjae, Hi. Wu. > > On Mon, Aug 27, 2012 at 12:01:14PM -0400, Namjae Jeon wrote: >> From: Namjae Jeon >> >> This patch is based on suggestion by Wu Fengguang: >> https://lkml.org/lkml/2011/8/19/19 >> >> kernel has mechanism to do writeback as per dirty_ratio and >> d

Re: [PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-09-10 Thread Fengguang Wu
Hi Namjae, On Mon, Aug 27, 2012 at 12:01:14PM -0400, Namjae Jeon wrote: > From: Namjae Jeon > > This patch is based on suggestion by Wu Fengguang: > https://lkml.org/lkml/2011/8/19/19 > > kernel has mechanism to do writeback as per dirty_ratio and dirty_background > ratio. It also maintains per

[PATCH 1/2] writeback: add dirty_background_time per bdi variable

2012-08-27 Thread Namjae Jeon
From: Namjae Jeon This patch is based on suggestion by Wu Fengguang: https://lkml.org/lkml/2011/8/19/19 kernel has mechanism to do writeback as per dirty_ratio and dirty_background ratio. It also maintains per task dirty rate limit to keep balance of dirty pages at any given instance by doing bd