[RFC v1] add new io-scheduler to use cgroup on high-speed device

2013-06-04 Thread Robin Dong
9300 54 371 test4 5806 87 393 Signed-off-by: Robin Dong Signed-off-by: Zhu Yanhai Cc: Tejun Heo Cc: Vivek Goyal Cc: Jens Axboe Cc: Tao Ma --- block/Kconfig.iosched | 13 + block/Makefile |1 + block/tpps-iosched.

[RFC v1] add new io-scheduler to use cgroup on high-speed device

2013-06-04 Thread Robin Dong
From: Robin Dong We want to use blkio.cgroup on high-speed device (like fusionio) for our mysql clusters. After testing different io-scheduler, we found that cfq is too slow and deadline can't run on cgroup. So we developed a new io-scheduler: tpps (Tiny Parallel Proportion Schedule

Re: [PATCH 2/2] md: modify dm_io() so it could return bios instead of submitting it

2012-11-13 Thread Robin Dong
ping 2012/9/20 Robin Dong : > From: Robin Dong > > When trying to modify flashcache to request based (current it's bio based), > we need > to make request from bios by ourselves, but dm_io() will submit these bios > directly, > so we propose to modify the dm_io()

Re: [PATCH 1/2] md: add new interface 'mk_rq' in target_type

2012-11-13 Thread Robin Dong
ping 2012/9/20 Robin Dong : > From: Robin Dong > > We are now trying to modify flashcache(https://github.com/facebook/flashcache) > to make it request based so that > we can let cfq io-controller control the bandwidth between different > io cgroups. > > A search in the dm

[PATCH 2/2 v5] block/throttle: Add IO submitted information in blkio.throttle

2012-11-02 Thread Robin Dong
From: Robin Dong Currently, if the IO is throttled by io-throttle, the system admin has no idea of the situation and can't report it to the real application user about that he/she has to do something. So this patch adds a new interface named blkio.throttle.io_submitted which exposes the n

[PATCH 1/2 v5] block/throttle: remove redundant type transition

2012-11-02 Thread Robin Dong
From: Robin Dong We don't need to convert tg to blkg and then convert it back in throtl_update_dispatch_stats(). Signed-off-by: Robin Dong --- block/blk-throttle.c |7 +++ 1 files changed, 3 insertions(+), 4 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c

Re: [PATCH 2/2 v4] block/throttle: Add IO submitted information in blkio.throttle

2012-10-27 Thread Robin Dong
2012/10/27 Vivek Goyal : > On Fri, Oct 26, 2012 at 12:47:48PM +0800, Robin Dong wrote: > > [..] >> @@ -1084,6 +1114,16 @@ static struct cftype throtl_files[] = { >> .private = offsetof(struct tg_stats_cpu, serviced), >> .read_seq_s

[PATCH 2/2 v4] block/throttle: Add IO submitted information in blkio.throttle

2012-10-25 Thread Robin Dong
From: Robin Dong Currently, if the IO is throttled by io-throttle, the system admin has no idea of the situation and can't report it to the real application user about that he/she has to do something. So this patch adds a new interface named blkio.throttle.io_submitted which exposes the n

[PATCH 1/2 v4] block/throttle: remove redundant type transition

2012-10-25 Thread Robin Dong
From: Robin Dong We don't need to convert tg to blkg and then convert it back in throtl_update_dispatch_stats(). Signed-off-by: Robin Dong --- block/blk-throttle.c |7 +++ 1 files changed, 3 insertions(+), 4 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c

[PATCH 2/2 v4] block/throttle: Add IO queued information in blkio.throttle

2012-10-18 Thread Robin Dong
From: Robin Dong Currently, if the IO is throttled by io-throttle, the system admin has no idea of the situation and can't report it to the real application user about that he/she has to do something. So this patch adds a new interface named blkio.throttle.io_queued which exposes the numb

[PATCH 1/2 v4] block/throttle: remove redundant type transition

2012-10-18 Thread Robin Dong
From: Robin Dong We don't need to convert tg to blkg and then convert it back in throtl_update_dispatch_stats(). Signed-off-by: Robin Dong --- block/blk-throttle.c |7 +++ 1 files changed, 3 insertions(+), 4 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c

[PATCH V3] block/throttle: Add IO throttled information in blkio.throttle

2012-10-08 Thread Robin Dong
From: Robin Dong Currently, if the IO is throttled by io-throttle, the SA has no idea of the situation and can't report it to the real application user about that he/she has to do something. So this patch adds a new interface named blkio.throttle.io_queued which indicates how many IO

[PATCH 2/2] md: modify dm_io() so it could return bios instead of submitting it

2012-09-19 Thread Robin Dong
From: Robin Dong When trying to modify flashcache to request based (current it's bio based), we need to make request from bios by ourselves, but dm_io() will submit these bios directly, so we propose to modify the dm_io() to return bios instead of submiting it. This could also improv

[PATCH 1/2] md: add new interface 'mk_rq' in target_type

2012-09-19 Thread Robin Dong
From: Robin Dong We are now trying to modify flashcache(https://github.com/facebook/flashcache) to make it request based so that we can let cfq io-controller control the bandwidth between different io cgroups. A search in the dm directory tells me that only multipath is a request based dm

[PATCH 2/2] md: modify dm_io() so it could return bios instead of submitting it

2012-09-02 Thread Robin Dong
From: Robin Dong When trying to modify flashcache to request based (current it's bio based), we need to make request from bios by ourselves, but dm_io() will submit these bios directly, so we propose to modify the dm_io() to return bios instead of submiting it. This could also improv

[PATCH 1/2] md: add new interface 'mk_rq' in target_type

2012-09-02 Thread Robin Dong
From: Robin Dong We are now trying to modify flashcache(https://github.com/facebook/flashcache) to make it request based so that we can let cfq io-controller control the bandwidth between different io cgroups. A search in the dm directory tells me that only multipath is a request based dm