On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer wrote:
> On Wed, Jun 10 2015 at 5:20pm -0400,
> Ming Lin wrote:
>
>> On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin wrote:
>> > On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
>> >> We need to test on large HW raid setups like a Netapp filer (or
On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer snit...@redhat.com wrote:
On Wed, Jun 10 2015 at 5:20pm -0400,
Ming Lin m...@kernel.org wrote:
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
We need to test on large HW
On Wed, 2015-06-10 at 15:06 -0700, Ming Lin wrote:
> On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer wrote:
> > On Wed, Jun 10 2015 at 5:20pm -0400,
> > Ming Lin wrote:
> >
> >> On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin wrote:
> >> > On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
> >> >>
On Wed, 2015-06-10 at 15:06 -0700, Ming Lin wrote:
On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer snit...@redhat.com wrote:
On Wed, Jun 10 2015 at 5:20pm -0400,
Ming Lin m...@kernel.org wrote:
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-06-04 at 17:06
On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer wrote:
> On Wed, Jun 10 2015 at 5:20pm -0400,
> Ming Lin wrote:
>
>> On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin wrote:
>> > On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
>> >> We need to test on large HW raid setups like a Netapp filer (or
On Wed, Jun 10 2015 at 5:20pm -0400,
Ming Lin wrote:
> On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin wrote:
> > On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
> >> We need to test on large HW raid setups like a Netapp filer (or even
> >> local SAS drives connected via some SAS controller).
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin wrote:
> On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
>> We need to test on large HW raid setups like a Netapp filer (or even
>> local SAS drives connected via some SAS controller). Like a 8+2 drive
>> RAID6 or 8+1 RAID5 setup. Testing with MD
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
We need to test on large HW raid setups like a Netapp filer (or even
local SAS drives connected via some SAS controller). Like a 8+2 drive
RAID6 or 8+1 RAID5 setup. Testing
On Wed, Jun 10 2015 at 5:20pm -0400,
Ming Lin m...@kernel.org wrote:
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
We need to test on large HW raid setups like a Netapp filer (or even
local SAS drives connected via
On Wed, Jun 10, 2015 at 2:46 PM, Mike Snitzer snit...@redhat.com wrote:
On Wed, Jun 10 2015 at 5:20pm -0400,
Ming Lin m...@kernel.org wrote:
On Mon, Jun 8, 2015 at 11:09 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
We need to test on large HW
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
> We need to test on large HW raid setups like a Netapp filer (or even
> local SAS drives connected via some SAS controller). Like a 8+2 drive
> RAID6 or 8+1 RAID5 setup. Testing with MD raid on JBOD setups with 8
> devices is also useful.
On Thu, 2015-06-04 at 17:06 -0400, Mike Snitzer wrote:
We need to test on large HW raid setups like a Netapp filer (or even
local SAS drives connected via some SAS controller). Like a 8+2 drive
RAID6 or 8+1 RAID5 setup. Testing with MD raid on JBOD setups with 8
devices is also useful. It
On Thu, Jun 4, 2015 at 5:06 PM, Mike Snitzer wrote:
> On Thu, Jun 04 2015 at 6:21pm -0400,
> Ming Lin wrote:
>
>> On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer wrote:
>> >
>> > We need to test on large HW raid setups like a Netapp filer (or even
>> > local SAS drives connected via some SAS
On Thu, Jun 04 2015 at 6:21pm -0400,
Ming Lin wrote:
> On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer wrote:
> >
> > We need to test on large HW raid setups like a Netapp filer (or even
> > local SAS drives connected via some SAS controller). Like a 8+2 drive
> > RAID6 or 8+1 RAID5 setup.
On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer wrote:
> On Tue, Jun 02 2015 at 4:59pm -0400,
> Ming Lin wrote:
>
>> On Sun, May 31, 2015 at 11:02 PM, Ming Lin wrote:
>> > On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
>> >> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
On Tue, Jun 02 2015 at 4:59pm -0400,
Ming Lin wrote:
> On Sun, May 31, 2015 at 11:02 PM, Ming Lin wrote:
> > On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
> >> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
> >> > Here are fio results of XFS on a DM stripped target
On Thu, Jun 4, 2015 at 5:06 PM, Mike Snitzer snit...@redhat.com wrote:
On Thu, Jun 04 2015 at 6:21pm -0400,
Ming Lin m...@kernel.org wrote:
On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer snit...@redhat.com wrote:
We need to test on large HW raid setups like a Netapp filer (or even
local
On Tue, Jun 02 2015 at 4:59pm -0400,
Ming Lin m...@kernel.org wrote:
On Sun, May 31, 2015 at 11:02 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM
On Thu, Jun 04 2015 at 6:21pm -0400,
Ming Lin m...@kernel.org wrote:
On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer snit...@redhat.com wrote:
We need to test on large HW raid setups like a Netapp filer (or even
local SAS drives connected via some SAS controller). Like a 8+2 drive
RAID6
On Thu, Jun 4, 2015 at 2:06 PM, Mike Snitzer snit...@redhat.com wrote:
On Tue, Jun 02 2015 at 4:59pm -0400,
Ming Lin m...@kernel.org wrote:
On Sun, May 31, 2015 at 11:02 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
On Wed, May 27, 2015 at
On Tue, 26 May 2015 16:34:14 +0100 Alasdair G Kergon wrote:
> On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
> > Now bio_add_page() can build lager bios.
> > And blk_queue_split() can split the bios in ->make_request() if needed.
>
> But why not try to make the bio the right size in
On Sun, May 31, 2015 at 11:02 PM, Ming Lin wrote:
> On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
>> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
>> > Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
>> > Does it make sense?
>>
>> To stripe
On Sun, May 31, 2015 at 11:02 PM, Ming Lin m...@kernel.org wrote:
On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make sense?
To stripe
On Tue, 26 May 2015 16:34:14 +0100 Alasdair G Kergon a...@redhat.com wrote:
On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
Now bio_add_page() can build lager bios.
And blk_queue_split() can split the bios in -make_request() if needed.
But why not try to make the bio the right
On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
> > Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
> > Does it make sense?
>
> To stripe across devices with different characteristics?
>
> Some
On Thu, 2015-05-28 at 01:36 +0100, Alasdair G Kergon wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make sense?
To stripe across devices with different characteristics?
Some suggestions.
On Fri, May 29 2015 at 3:05P -0400,
Ming Lin wrote:
> On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon wrote:
> > On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
> >> Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
> >> Does it make sense?
> >
> > To stripe
On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon wrote:
> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
>> Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
>> Does it make sense?
>
> To stripe across devices with different characteristics?
>
> Some
On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon a...@redhat.com wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make sense?
To stripe across devices with different characteristics?
Some
On Fri, May 29 2015 at 3:05P -0400,
Ming Lin m...@kernel.org wrote:
On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon a...@redhat.com wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make
On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon wrote:
> On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
>> Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
>> Does it make sense?
>
> To stripe across devices with different characteristics?
I intended to test
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
> Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
> Does it make sense?
To stripe across devices with different characteristics?
Some suggestions.
Prepare 3 kernels.
O - Old kernel.
M - Old kernel with
On Tue, May 26, 2015 at 9:04 AM, Mike Snitzer wrote:
> On Tue, May 26 2015 at 11:02am -0400,
> Ming Lin wrote:
>
>> On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer wrote:
>> > On Fri, May 22 2015 at 2:18pm -0400,
>> > Ming Lin wrote:
>> >
>> >> From: Kent Overstreet
>> >>
>> >> The way the
On Wed, May 27, 2015 at 01:40:22AM +0100, Alasdair G Kergon wrote:
> It does sometimes make a significant difference to device-mapper stacks.
> DM only uses it for performance reasons - it can already split bios when
> it needs to. I tried to remove merge_bvec_fn from DM several years ago but
>
On Wed, May 27, 2015 at 01:40:22AM +0100, Alasdair G Kergon wrote:
It does sometimes make a significant difference to device-mapper stacks.
DM only uses it for performance reasons - it can already split bios when
it needs to. I tried to remove merge_bvec_fn from DM several years ago but
On Tue, May 26, 2015 at 9:04 AM, Mike Snitzer snit...@redhat.com wrote:
On Tue, May 26 2015 at 11:02am -0400,
Ming Lin m...@kernel.org wrote:
On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer snit...@redhat.com wrote:
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin m...@kernel.org wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make sense?
To stripe across devices with different characteristics?
Some suggestions.
Prepare 3 kernels.
O - Old kernel.
M - Old kernel with
On Wed, May 27, 2015 at 5:36 PM, Alasdair G Kergon a...@redhat.com wrote:
On Wed, May 27, 2015 at 04:42:44PM -0700, Ming Lin wrote:
Here are fio results of XFS on a DM stripped target with 2 SSDs + 1 HDD.
Does it make sense?
To stripe across devices with different characteristics?
I intended
On Wed, May 27, 2015 at 09:06:40AM +1000, Neil Brown wrote:
> Because we don't know what the "right" size is. And the "right" size can
> change when array reconfiguration happens.
In certain configurations today, device-mapper does report back a sensible
maximum bio size smaller than would
On Tue, 26 May 2015 16:34:14 +0100 Alasdair G Kergon wrote:
> On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
> > Now bio_add_page() can build lager bios.
> > And blk_queue_split() can split the bios in ->make_request() if needed.
>
> But why not try to make the bio the right size in
On Tue, May 26, 2015 at 9:04 AM, Mike Snitzer wrote:
> On Tue, May 26 2015 at 11:02am -0400,
> Ming Lin wrote:
>
>> On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer wrote:
>> > On Fri, May 22 2015 at 2:18pm -0400,
>> > Ming Lin wrote:
>> >
>> >> From: Kent Overstreet
>> >>
>> >> The way the
On Tue, May 26 2015 at 11:02am -0400,
Ming Lin wrote:
> On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer wrote:
> > On Fri, May 22 2015 at 2:18pm -0400,
> > Ming Lin wrote:
> >
> >> From: Kent Overstreet
> >>
> >> The way the block layer is currently written, it goes to great lengths
> >> to
On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
> Now bio_add_page() can build lager bios.
> And blk_queue_split() can split the bios in ->make_request() if needed.
But why not try to make the bio the right size in the first place so you
don't have to incur the performance impact of
On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer wrote:
> On Fri, May 22 2015 at 2:18pm -0400,
> Ming Lin wrote:
>
>> From: Kent Overstreet
>>
>> The way the block layer is currently written, it goes to great lengths
>> to avoid having to split bios; upper layer code (such as bio_add_page())
>>
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin wrote:
> From: Kent Overstreet
>
> The way the block layer is currently written, it goes to great lengths
> to avoid having to split bios; upper layer code (such as bio_add_page())
> checks what the underlying device can handle and tries to always
On Wed, May 27, 2015 at 09:06:40AM +1000, Neil Brown wrote:
Because we don't know what the right size is. And the right size can
change when array reconfiguration happens.
In certain configurations today, device-mapper does report back a sensible
maximum bio size smaller than would otherwise
On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
Now bio_add_page() can build lager bios.
And blk_queue_split() can split the bios in -make_request() if needed.
But why not try to make the bio the right size in the first place so you
don't have to incur the performance impact of
On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer snit...@redhat.com wrote:
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin m...@kernel.org wrote:
From: Kent Overstreet kent.overstr...@gmail.com
The way the block layer is currently written, it goes to great lengths
to avoid having to split bios;
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin m...@kernel.org wrote:
From: Kent Overstreet kent.overstr...@gmail.com
The way the block layer is currently written, it goes to great lengths
to avoid having to split bios; upper layer code (such as bio_add_page())
checks what the underlying
On Tue, 26 May 2015 16:34:14 +0100 Alasdair G Kergon a...@redhat.com wrote:
On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote:
Now bio_add_page() can build lager bios.
And blk_queue_split() can split the bios in -make_request() if needed.
But why not try to make the bio the right
On Tue, May 26 2015 at 11:02am -0400,
Ming Lin m...@kernel.org wrote:
On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer snit...@redhat.com wrote:
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin m...@kernel.org wrote:
From: Kent Overstreet kent.overstr...@gmail.com
The way the block layer
On Tue, May 26, 2015 at 9:04 AM, Mike Snitzer snit...@redhat.com wrote:
On Tue, May 26 2015 at 11:02am -0400,
Ming Lin m...@kernel.org wrote:
On Tue, May 26, 2015 at 7:36 AM, Mike Snitzer snit...@redhat.com wrote:
On Fri, May 22 2015 at 2:18pm -0400,
Ming Lin m...@kernel.org wrote:
On Fri, 22 May 2015 11:18:33 -0700 Ming Lin wrote:
> From: Kent Overstreet
>
> The way the block layer is currently written, it goes to great lengths
> to avoid having to split bios; upper layer code (such as bio_add_page())
> checks what the underlying device can handle and tries to always
On Fri, 22 May 2015 11:18:33 -0700 Ming Lin m...@kernel.org wrote:
From: Kent Overstreet kent.overstr...@gmail.com
The way the block layer is currently written, it goes to great lengths
to avoid having to split bios; upper layer code (such as bio_add_page())
checks what the underlying
From: Kent Overstreet
The way the block layer is currently written, it goes to great lengths
to avoid having to split bios; upper layer code (such as bio_add_page())
checks what the underlying device can handle and tries to always create
bios that don't need to be split.
But this approach
From: Kent Overstreet kent.overstr...@gmail.com
The way the block layer is currently written, it goes to great lengths
to avoid having to split bios; upper layer code (such as bio_add_page())
checks what the underlying device can handle and tries to always create
bios that don't need to be split.
56 matches
Mail list logo