zfs discuss - Jun 2010 - OCZ Vertex 2 Pro performance numbers

Now the test for the Vertex 2 Pro. This was fun.
For more explanation please see the thread "Crucial RealSSD C300 and cache
flush?"
This time I made sure the device is attached via 3GBit SATA. This is also
only a short test. I''ll retest after some weeks of usage.

cache enabled, 32 buffers, 64k blocks
linear write, random data: 96 MB/s
linear read, random data: 206 MB/s
linear write, zero data: 234 MB/s
linear read, zero data: 255 MB/s
random write, random data: 84 MB/s
random read, random data: 180 MB/s
random write, zero data: 224 MB/s
randow read, zero data: 190 MB/s

cache enabled, 32 buffers, 4k blocks
linear write, random data: 93 MB/s
linear read, random data: 138 MB/s
linear write, zero data: 113 MB/s
linear read, zero data: 141 MB/s
random write, random data: 41 MB/s (10300 ops/s)
random read, random data: 76 MB/s (19000 ops/s)
random write, zero data: 54 MB/s (13800 ops/s)
random read, zero data: 91 MB/s (22800 ops/s)


cache enabled, 1 buffer, 4k blocks
linear write, random data: 62 MB/s (15700 ops/s)
linear read, random data: 32 MB/s (8000 ops/s)
linear write, zero data: 64 MB/s (16100 ops/s)
linear read, zero data: 45 MB/s (11300 ops/s)
random write, random data: 14 MB/s (3400 ops/s)
random read, random data: 22 MB/s (5600 ops/s)
random write, zero data: 19 MB/s (4500 ops/s)
random read, zero data: 21 MB/s (5100 ops/s)

cache enabled, 1 buffer, 4k blocks, with cache flushes:
linear write, random data, flush after every write: 5700 ops/s
linear write, zero data, flush after every write: 5700 ops/s
linear write, random data, flush after every 4th write: 8500 ops/s
linear write, zero data, flush after every 4th write: 8500 ops/s

Some remarks:

The random op numbers have to be read with care:
 - reading occurs in the same order as the writing before
 - the ops are not aligned to any specific boundary

The device also passed the write-loss-test: after 5 repeats no
data has been lost.

It doesn''t make any difference if the cache is enabled or disabled, so
it might be worth to tune zfs to not issue cache flushes.

Conclusion: This device will make an excellent slog device. I''ll order
them today ;)

--Arne

>
>
> Conclusion: This device will make an excellent slog device. I''ll
order
> them today ;)
>
>
I have one and i love it...I sliced it though, used 9 gb for ZIL and the
rest for L2ARC (my server is on a smallish network with about 10 clients)

It made a huge difference in NFS performance and other stuff as well (for
instance, doing something like du will run a TON faster than before)

For the money, it''s a GREAT deal.  I am very impressed


> --Arne
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20100625/21083f47/attachment.html>

On 25 Jun 2010, at 15:23, Thomas Burgess <wonslung at
gmail.com<mailto:wonslung at gmail.com>> wrote:


Conclusion: This device will make an excellent slog device. I''ll order
them today ;)


I have one and i love it...I sliced it though, used 9 gb for ZIL and the rest
for L2ARC (my server is on a smallish network with about 10 clients)

It made a huge difference in NFS performance and other stuff as well (for
instance, doing something like du will run a TON faster than before)

For the money, it''s a GREAT deal.  I am very impressed


--Arne
_______________________________________________
zfs-discuss mailing list
<mailto:zfs-discuss at opensolaris.org>zfs-discuss at
opensolaris.org<mailto:zfs-discuss at opensolaris.org>
<http://mail.opensolaris.org/mailman/listinfo/zfs-discuss>http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

<ATT00001..c>

Would be great if someone could test the SLC EX version of this drive.

---Przem
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20100625/801e6f8e/attachment.html>

>From: Arne Jansen
>Sent: Friday, June 25, 2010 3:21 AM
>
>Now the test for the Vertex 2 Pro. This was fun.
>For more explanation please see the thread "Crucial RealSSD C300 and
cache
>flush?"
>This time I made sure the device is attached via 3GBit SATA. This is also
only a
>short test. I''ll retest after some weeks of usage.
>
>cache enabled, 32 buffers, 64k blocks
>linear write, random data: 96 MB/s
>linear read, random data: 206 MB/s
>linear write, zero data: 234 MB/s
>linear read, zero data: 255 MB/s
>random write, random data: 84 MB/s
>random read, random data: 180 MB/s
>random write, zero data: 224 MB/s
>randow read, zero data: 190 MB/s
>
>cache enabled, 32 buffers, 4k blocks
>linear write, random data: 93 MB/s
>linear read, random data: 138 MB/s
>linear write, zero data: 113 MB/s
>linear read, zero data: 141 MB/s
>random write, random data: 41 MB/s (10300 ops/s) random read, random data:
>76 MB/s (19000 ops/s) random write, zero data: 54 MB/s (13800 ops/s) random
>read, zero data: 91 MB/s (22800 ops/s)
>
>
>cache enabled, 1 buffer, 4k blocks
>linear write, random data: 62 MB/s (15700 ops/s) linear read, random data:
32
>MB/s (8000 ops/s) linear write, zero data: 64 MB/s (16100 ops/s) linear
read, zero
>data: 45 MB/s (11300 ops/s) random write, random data: 14 MB/s (3400 ops/s)
>random read, random data: 22 MB/s (5600 ops/s) random write, zero data: 19
>MB/s (4500 ops/s) random read, zero data: 21 MB/s (5100 ops/s)
>
>cache enabled, 1 buffer, 4k blocks, with cache flushes:
>linear write, random data, flush after every write: 5700 ops/s linear
write, zero
>data, flush after every write: 5700 ops/s linear write, random data, flush
after
>every 4th write: 8500 ops/s linear write, zero data, flush after every 4th
write:
>8500 ops/s
>
>
>Some remarks:
>
>The random op numbers have to be read with care:
> - reading occurs in the same order as the writing before
> - the ops are not aligned to any specific boundary
>
>The device also passed the write-loss-test: after 5 repeats no data has
been lost.
>
>It doesn''t make any difference if the cache is enabled or disabled,
so it
might be
>worth to tune zfs to not issue cache flushes.
>
>Conclusion: This device will make an excellent slog device. I''ll
order them
today ;)
>
>--Arne
Arne, thanks for doing these tests, they are great to see.  

Is this the one
(http://www.ocztechnology.com/products/solid-state-drives/2-5--sata-ii/maxim
um-performance-enterprise-solid-state-drives/ocz-vertex-2-pro-series-sata-ii
-2-5--ssd-.html) with the built in supercap? 

Geoff

Geoff Nordli wrote:
> 
> Is this the one
>
(http://www.ocztechnology.com/products/solid-state-drives/2-5--sata-ii/maxim
>
um-performance-enterprise-solid-state-drives/ocz-vertex-2-pro-series-sata-ii
> -2-5--ssd-.html) with the built in supercap? 
> 
Yes.
> Geoff

On Jun 26, 2010, at 02:09, Arne Jansen wrote:
> Geoff Nordli wrote:
>> Is this the one
>>
(http://www.ocztechnology.com/products/solid-state-drives/2-5--sata-ii/maxim
>> um-performance-enterprise-solid-state-drives/ocz-vertex-2-pro- 
>> series-sata-ii
>> -2-5--ssd-.html) with the built in supercap?
>
> Yes.
Crickey. Who''s the genius who thinks of these URLs?

On 6/26/10 9:47 AM -0400 David Magda wrote:
> Crickey. Who''s the genius who thinks of these URLs?
SEOs

Any duration limit on the supercap? How long can it sustain the data?

Thanks.

Fred

-----Original Message-----
From: zfs-discuss-bounces at opensolaris.org [mailto:zfs-discuss-bounces at
opensolaris.org] On Behalf Of David Magda
Sent: ???, ?? 26, 2010 21:48
To: Arne Jansen
Cc: ''OpenSolaris ZFS discuss''
Subject: Re: [zfs-discuss] OCZ Vertex 2 Pro performance numbers


On Jun 26, 2010, at 02:09, Arne Jansen wrote:
> Geoff Nordli wrote:
>> Is this the one
>>
(http://www.ocztechnology.com/products/solid-state-drives/2-5--sata-ii/maxim
>> um-performance-enterprise-solid-state-drives/ocz-vertex-2-pro- 
>> series-sata-ii
>> -2-5--ssd-.html) with the built in supercap?
>
> Yes.
Crickey. Who''s the genius who thinks of these URLs?
_______________________________________________
zfs-discuss mailing list
zfs-discuss at opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

On Wed, 30 Jun 2010, Fred Liu wrote:
> Any duration limit on the supercap? How long can it sustain the data?
A supercap on a SSD drive only needs to sustain the data until it has 
been saved (perhaps 10 milliseconds).  It is different than a RAID 
array battery.

Bob
-- 
Bob Friesenhahn
bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/

See. Thanks.
Does it have the hardware functionality to detect the power outage and do force
cache flush when the cache is enabled?
Any more detailed info about the capacity (farad) of this supercap and how long
one discharge will be?

Thanks.

Fred

-----Original Message-----
From: Bob Friesenhahn [mailto:bfriesen at simple.dallas.tx.us] 
Sent: ???, ?? 01, 2010 10:01
To: Fred Liu
Cc: ''OpenSolaris ZFS discuss''
Subject: Re: [zfs-discuss] OCZ Vertex 2 Pro performance numbers

On Wed, 30 Jun 2010, Fred Liu wrote:
> Any duration limit on the supercap? How long can it sustain the data?
A supercap on a SSD drive only needs to sustain the data until it has 
been saved (perhaps 10 milliseconds).  It is different than a RAID 
array battery.

Bob
-- 
Bob Friesenhahn
bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/

On 6/30/2010 7:17 PM, Fred Liu wrote:
> See. Thanks.
> Does it have the hardware functionality to detect the power outage and do
force cache flush when the cache is enabled?
> Any more detailed info about the capacity (farad) of this supercap and how
long one discharge will be?
>
> Thanks.
>
> Fred
>
>   
I don''t think it matters the actual size of the supercap. As Bob said,
it only needs to be sized to be big enough to allow all on-board DRAM to
be flushed out to Flash. How big that should be is easily determined by
the manufacturer, and they''d be grossly negligent if it wasn''t
at least
that size. Any capacity beyond that needed to do a single full flush is
excess, so I would hazard a guess that the supercap is "just big enough,
and no more". That is, just enough to power the SSD for the partial
second or so it takes to flush to flash. I don''t think we need to worry
how big that actually is.

Answering your second question first - my reading of the info is that it
will force a cache flush (if the cache is enabled) upon loss of power
under any circumstance. That is, it will flush the cache in both a
controlled power-down (regardless of whether the OS says to do so) and
in an immediate power loss. All this is in the SSD''s firmware.

-Erik

> -----Original Message-----
> From: Bob Friesenhahn [mailto:bfriesen at simple.dallas.tx.us] 
> Sent: ???, ?? 01, 2010 10:01
> To: Fred Liu
> Cc: ''OpenSolaris ZFS discuss''
> Subject: Re: [zfs-discuss] OCZ Vertex 2 Pro performance numbers
>
> On Wed, 30 Jun 2010, Fred Liu wrote:
>
>   
>> Any duration limit on the supercap? How long can it sustain the data?
>>     
> A supercap on a SSD drive only needs to sustain the data until it has 
> been saved (perhaps 10 milliseconds).  It is different than a RAID 
> array battery.
>
> Bob
>   

-- 
Erik Trimble
Java System Support
Mailstop:  usca22-123
Phone:  x17195
Santa Clara, CA
Timezone: US/Pacific (GMT-0800)

> -----Original Message-----
> From: Erik Trimble [mailto:erik.trimble at oracle.com]
> Sent: ???, ?? 01, 2010 11:45
> To: Fred Liu
> Cc: Bob Friesenhahn; ''OpenSolaris ZFS discuss''
> Subject: Re: [zfs-discuss] OCZ Vertex 2 Pro performance numbers
> 
> On 6/30/2010 7:17 PM, Fred Liu wrote:
> > See. Thanks.
> > Does it have the hardware functionality to detect the power outage
> and do force cache flush when the cache is enabled?
> > Any more detailed info about the capacity (farad) of this supercap
> and how long one discharge will be?
> >
> > Thanks.
> >
> > Fred
> >
> >
> 
> I don''t think it matters the actual size of the supercap. As Bob
said,
> it only needs to be sized to be big enough to allow all on-board DRAM
> to
> be flushed out to Flash. How big that should be is easily determined by
> the manufacturer, and they''d be grossly negligent if it
wasn''t at least
> that size. Any capacity beyond that needed to do a single full flush is
> excess, so I would hazard a guess that the supercap is "just big
enough,
> and no more". That is, just enough to power the SSD for the partial
> second or so it takes to flush to flash. I don''t think we need to
worry
> how big that actually is.
Understand and agree. It is sort of picky to ask this without
manufacturer''s help ;-)
> 
> Answering your second question first - my reading of the info is that
> it
> will force a cache flush (if the cache is enabled) upon loss of power
> under any circumstance. That is, it will flush the cache in both a
> controlled power-down (regardless of whether the OS says to do so) and
> in an immediate power loss. All this is in the SSD''s firmware.
That is exactly what I expect. Little bit broadly speaking, it is sort of
ambiguous in
aspect of cache flush in all the HDDs. Is it OS-controlled or
firmware-controlled or both?
At least in this case, I have got what I expect. But what about for all(generic)
the HDDs?

Thanks.

Fred
> 
> -Erik
> 
> 
> > -----Original Message-----
> > From: Bob Friesenhahn [mailto:bfriesen at simple.dallas.tx.us]
> > Sent: ???, ?? 01, 2010 10:01
> > To: Fred Liu
> > Cc: ''OpenSolaris ZFS discuss''
> > Subject: Re: [zfs-discuss] OCZ Vertex 2 Pro performance numbers
> >
> > On Wed, 30 Jun 2010, Fred Liu wrote:
> >
> >
> >> Any duration limit on the supercap? How long can it sustain the
data?
> >>
> > A supercap on a SSD drive only needs to sustain the data until it has
> > been saved (perhaps 10 milliseconds).  It is different than a RAID
> > array battery.
> >
> > Bob
> >
> 
> 
> --
> Erik Trimble
> Java System Support
> Mailstop:  usca22-123
> Phone:  x17195
> Santa Clara, CA
> Timezone: US/Pacific (GMT-0800)
>

(I am aware I am replying to an old post...)

Arne Jansen <sensille <at> gmx.net> writes:
> 
> Now the test for the Vertex 2 Pro. This was fun.
> For more explanation please see the thread "Crucial RealSSD C300 and
cache
> flush?"
> This time I made sure the device is attached via 3GBit SATA. This is also
> only a short test. I''ll retest after some weeks of usage.
> 
> cache enabled, 32 buffers, 64k blocks
> linear write, random data: 96 MB/s
> linear read, random data: 206 MB/s
> linear write, zero data: 234 MB/s
> linear read, zero data: 255 MB/s
This discrepancy between tests with random data and zero data is puzzling
to me. Does this suggest that the SSD does transparent compression between
its Sandforce SF-1500 controller and the NAND flash chips?
> cache enabled, 32 buffers, 4k blocks
> random write, random data: 41 MB/s (10300 ops/s)
> random read, random data: 76 MB/s (19000 ops/s)
> random write, zero data: 54 MB/s (13800 ops/s)
> random read, zero data: 91 MB/s (22800 ops/s)
These IOPS numbers are significantly below the 50000 IOPS announced
by OCZ. But I supposed this is due to your benchmark tool not aligning
the ops to 4K boundaries, correct?

-mrb

Marc Bevand <m.bevand <at> gmail.com>
writes:
> 
> This discrepancy between tests with random data and zero data is puzzling
> to me. Does this suggest that the SSD does transparent compression between
> its Sandforce SF-1500 controller and the NAND flash chips?
Replying to myself: yes, SF-1500 does transparent deduplication
and compression to reduce write-amplification. Wow.

http://www.semiaccurate.com/2010/05/03/sandforce-ssds-break-tpc-c-records/

-mrb