Date:	Tue, 29 Jun 2010 16:35:46 +0200
From:	Nebojsa Trpkovic <trx.lists@...il.com>
To:	tytso@....edu
CC:	linux-ext4@...r.kernel.org
Subject: Re: Ext4 on SSD Intel X25-M

> On 06/29/10 15:56, tytso@....edu wrote:
> On Sun, Jun 27, 2010 at 07:47:46PM +0200, Nebojsa Trpkovic wrote:
>> My best guess is that host itself uses a lot of optimisation to reduce
>> writing to NAND itself.
> 
> Possible, although if the counter is defined as "host writes", that
> should be before the NAND writes, since "host writes" would expect
> means the actual write commands coming from the host -- i.e., coming
> incoming SATA write commands.

That's true. Maybe Intel gave wrong name to that variable, as they are
trying to keep track of NAND wear, not SATA write commands. Then again,
maybe it's just amount of SATA writes.

>> Besides that, I've noticed that my commit=100 mount option helps also.
>> Changing (just for testing) to something realy big, like commit=9000,
>> gave even further improvement, but not worth staying with risk of losing
>> (that much) data. It seems that ext4 writes a lot to filesystem, but
>> many of those writes are overwrites. If we flush them to host just once
>> in 100 seconds, we're getting a lot of saving.
> 
> What metric are you using when you say that this "helps"?  The ext4
> measurement, the SSD counter which you are using, or both?

I've made graph with rrdtool to track both lifetime_write_kbytes and
host writes from SSD. It looks like this:
http://img130.imageshack.us/img130/6905/systemg29year.png
with lifetime_writes_kbytes decreasing on unclean shutdown.
"host writes" counter read from SSD starts at bigger value because I've
done some testing with SSD before I've made sda1 and formated it.
Since then, trend is quite obvious. Differences in line slopes are due
different type of usage from time to time.

>> As I wanted to make even my swap TRIMable, I've put it in the file on
>> ext4 instead of separate partition. I've made it using dd with seek=500
>> bs=1M options. ext4's lifetime_write_kbytes increased by 500MB, and host
>> writes did not incrase at all, even after 100 seconds. Ok, I know that
>> ext4 did not write 500MB of data to filesystem, but this is one more
>> thing why one should not trust lifetime_write_kbytes.
> 
>> So, the moral of my story would be not to trust lifetime_write_kbytes,
>> but to read host writes from SSD.
> 
> If you wrote 500MB to a swap file in ext4 using dd, why are you sure
> ext4 didn't write 500MB of data to the disk?  In fact, this would
> imply to me that that your "host writes" shouldn't be trusted.

I've used dd with bs=1M, seek=500 and count=1 option, making file of
500MB but writting just last megabyte of it to avoid unnecessary SSD
writes during swap-file cration.

>> I noticed that Intel's Solid State Drive Toolbox software (running in
>> Windows) gives the amount of Host Lifetime Writes that equales to
>> S.M.A.R.T attribute 225 (Load_Cycle_Count) multiplied with 32MB.
>> That's the way I track it in Linux.
> 
> According to the S.M.A.R.T startd, Load_cycle_count is supposed to
> mean the number of times the platter has spun up and spin down.  It's
> not clear what it means for SDD's, so it may be that they have reused
> it for some other purpose.  However, it would be surprising to me that
> it was just host lifetime writes divided by 32MB.  It may be that you
> have noticed this correlation in Windows because Windows is very
> "chunky" in how it does its writes.

I've found out about that S.M.A.R.T value by constant comparing of "host
writes" that I see in Intel's SSD Toolbox and my calculated graph of
"host writes" (32MB * SMART value 225).
Everytime I reboot my box and boot Windows 7 residing on external HDD, I
start Intel SSD Toolbox that gives me "host writes" amount using nice
looking GUI. Whenever I rebooted in Windows, I've seen those values to
match. I guess that Intel just missuses S.M.A.R.T. value not needed for
theier SSD violating S.M.A.R.T. standard, but at least, I can get that
info in Linux without Intel SSD toolbox.

> However, if you write 500MB to a file in ext4 using dd, and ext4's
> lifetime_write_kbytes in /sysfs went up by 500MB, but the
> Load_Cycle_Count attribute did not go up, then I would deduce from
> that that your interpretation of Load_Cycle_Count is probably not
> correct...

I've explained this before. Maybe I'm wrong, but in my opinion

dd if=/dev/zero of=somefile bs=1M seek=500 count=1

should make 500MB file with just 1MB writen to disk. Even df should not
register absence of 500MB of free space.

Nebojsa
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists