OCZ Octane 512 GB and Indilinx Everest vs Crucial M4 512 GB - BeHardware
>> SSD

Written by Marc Prieur

Published on January 10, 2012

URL: http://www.behardware.com/art/lire/849/

Page 1

OCZ Octane and Indilinx Everest

Three years after the launch of the OCZ Vertex, OCZ and Indilinx have launched a new generation of SSD, the OCZ Octane range. While the two companies were previously partners, following OCZ’s buyout of Indilinx last year, they now form the same entity. From simply assembling the parts, OCZ has used 2011 to turn itself into a company that is capable of elaborating an SSD from A to Z, with the Octane the first in a long line of products.

A new controller
A new Indilinx SATA 6G controller is used on the OCZ Octane, the Everest. Several manufacturers such as Marvell or SandForce have been offering such a chip for some time, but the Indilinx solution was considerably delayed. In 2009, the Indilinx Barefoot was a precursor of high performance MLC SSDs and its SATA 6G successor, the Jet Stream, was initially slated for 2010 before being pushed back. In the end we had to wait for the end of the year to see Everest arrive.

Everest claims a high level of performance:

- sequential reads of 520 MB/s
- sequential writes of 410 MB/s
- random 4KB reads at 30K IOPS
- random 4KB writes at 22K IOPS

The controller is made up of an ARM type dual core CPU with a clock of up to 275 MHz and 196 KB of SRAM (128 KB for the firmware, 64 KB for data). It will have a 512 MB DDR2/3 DRAM cache running at 400 MHz and be able to address up to 1 TB of flash on 8 channels. This flash memory can be either MLC or SLC, with an asynchronous bus, ONFi or Toggle Mode. TRIM support is included of course, as is handling of wear and BCH ECC error correction enabling 70 bits capability per sector.
The OCZ Octane range
OCZ is going to be marketing several Indilinx Everest SSDs. From fastest to slowest we have the:

- OCZ Octane: SATA 6G, synchronous MLC
- OCZ Petrol: SATA 6G, asynchronous MLC
- OCZ Octane-S2: SATA 3G, asynchronous MLC memory

While the SSD capacity doesn’t affect read performance in the Octane range, with the exception of the 1 TB version, write performance is however very much influenced by capacity. This isn’t a new development in itself and such variation exists with most controllers, though on some Marvell SSDs, write performance won’t go any higher than 256 GB.
The OCZ Octane 512 GB

OCZ supplied us with the Octane 512 GB for this test. Inside the SSD you’ll find the new Indilinx IDX300 controller, accompanied by 16 (8 on either side of the PCB) MLC flash chips made by IMFT (Intels joint venture with Micron) as well as two Micron DRAM chips to make up the SSD’s 512 MB cache.

How does this SSD do in practice? As SSD performance is linked to capacity, we also obtained a Crucial M4 512 GB SSD so as to benchmark our Octane against an equivalent capacity model. We largely followed the protocol introduced in this article.

Page 2
Sequential and random throughputs

Synthetic performance – Sequential throughputs
We started with sequential writes, measured using IOMeter

The sequential speeds are very good, with reads of over 500 MB/s, the same as with the M4, and writes of 376 MB/s, which is significantly better than the M4 512 GB which doesn't do any better than the 256 GB version.
Synthetic performance – Random throughputs
Still with IOMeter, we measured performance during 4 KB random accesses across the whole SSD. These readings were carried out with a variable number of simultaneous accesses, going from 1 to 32, allowing us to illustrate the SSD’s ability to deal with these simultaneous accesses. While it’s definitely worth checking this out, we shouldn’t lose sight of the fact that in standard usage, the level of simultaneous accesses is somewhere between 1 and 4.

[ IOPS ]  [ MB/s ]

With reads, the OCZ Octane does better than the Crucial M4 with a single command, but the M4 has the advantage as of two commands, extending its lead after that.

[ IOPS ]  [ MB/s ]

When it comes to writes, the difference between the two is even more noticeable, with the Octane hitting a ceiling at around 20,000 accesses per second, against over 60,000 for the Crucial M4.

Page 3
Practical performance – File copying and applications

Practical performance – Files
Moving onto the practical tests, we started by looking at write and read speeds for various groups of files. These groups were composed as follows:

- Extra large: 731.17 MB on average
- Large files: 5.2 MB on average
- Medium sized files: 800.88 KB on average
- Small files: 48.78 KB on average

We used a RamDisk as the source or the target for reads or writes on the SSD. Given the speed of recent hard drives and so as to obtain results that are less subject to variation, we used Robocopy with some in-house software that allows us to carry the tests out continuously.

[ Reads ]  [ Writes ]

The two SSDs were pretty similar for reads. The Octane was faster on small files and the M4 was faster with the rest. The OCZ SSD stands out in writes however, especially with small files where it’s almost twice as fast!
Practical performance – Applications
Next we carried out the purely practical tests, namely various timed operations after installation of Windows 7 64-bit on each of the HDDs:

- Booting Windows 7
- Booting 3D Studio Max 2011
- Booting 3D Studio Max 2011 + Visual Studio 2010 + Bibble Pro 5
- Rescan of Ogre source code in Visual Studio 2010
- Regeneration of thumbnails from a directory of 48 RAW files in Bibble 5 Pro
- Launch of Battlefield 3
- Launch of a level in Battlefield 3

Note these timings are not comparable to those of previous tests. Firstly, the processor was overclocked to 4.5 GHz here. Secondly, for Windows 7 we now take a reading of the time between the start of the Windows boot (following the boot menu that can be accessed using F8) and the appearance of the desktop.

For 3d Studio Max 2011 we measure the time between launch and the appearance of the tips window, while the multi-application boot is carried out using a batch. The source code for the 3D Ogre engine is used in Visual Studio 2010, while a repertory containing 48 RAW files from a 5D mark II serves as the basis for the test in Bibble 5 Pro. In Battlefield 3 we measured the launch time of the game from the validation of the Origin password and the start of the introduction video, and the load of a level between validation of the resumption of the campaign (mission 7 – Thunder Run) and the appearance of the image on screen.

For information we also included the performance data for a Hitachi 7K3000 hard drive.

Windows takes 10 seconds to boot with both the OCZ Octane and the Crucial M4. While OCZ is highlighting its 'Fast boot' technology that is supposed to allow the Octane to achieve better results in this domain, it is in fact the M4 that is slightly faster.

The M4 is also fastest to boot 3d Studio Max and with the multi application boot. Compared to booting 3ds alone, launching two other applications at the same time only slows it down by 0.1s, as against 0.5s on the Octane. Of course, this type of situation only comes up quite rarely but it does allow us to see the big advantage SSDs have over hard drives: on the SSDs every task is executed more quickly and performance isn’t really affected when we’re doing something else at the same time, while on the 7K3000, launching Bibble and VS 2010 at the same time as 3ds takes 22.5 seconds longer than launching 3ds on its own.

Rescanning the Ogre source code is slightly faster on the Octane than on the M4, while it’s the other way round for the regeneration of thumbnails in Bibble. Lastly, in Battlefield 3 the M4 has the advantage again, but here it’s slight in comparison to the gains any SSD will give over a hard drive. Note however that not all games benefit as much from the use of SSDs and given the size of games and the cost per GB of SSDs, it is worth thinking twice before you splash the cash!

Page 4
Performance over time & TRIM

Performance over time & TRIM
As we’ve already mentioned on several occasions, an SSD’s performance levels can deteriorate over time. There are several reasons for this, the first being structural: a hard drive can read, write (unwritten space) or rewrite (space that has previously been written to) data by 4 KB packages. With flash memory you can only read, write or rewrite by 4 KB packages, 4 KB and … 512 KB (or even 8 KB, 8 KB and 2048 KB for a 25nm 8 GB flash chip).

When a space that’s already occupied by a file has to be rewritten, there can be an impact on performance! Worse still, if the file has been deleted by the OS and the OS doesn’t support TRIM, the SSD then doesn’t know that the file has been deleted and acts as if it had to rewrite the data rather than write it. This problem is however resolved with TRIM, as the OS then tells the SSD that the space is to be considered as unwritten.

This structural issue is accentuated by the existence of optimisations within SSDs aimed at improving random write performance and write amplification. To keep it simple, when a request to write data randomly to an SSD is received, the SSD writes it sequentially to the flash memory, making sure that the addresses in its internal allocation table correspond to the addresses known by the OS (the LBAs) and the corresponding flash pages. For such a mechanism to be efficient, the blocks of flash memory do have to be available, which can be problematic without TRIM support.

What does all this translate to in practice? Testing an SSD’s wear isn’t easy to do but we have evaluated how the SSDs do when faced with an extreme situation. This time, before taking any performance reading, all accessible space is filled with data written sequentially. The only flash cells considered as unwritten by the SSD are those set aside for overprovisioning, something that shouldn’t occur in a system that frees up cells correctly via TRIM.

We then took several performance measures after putting our SSDs through the following stages:
1. Filling the SSD
2. 20mn of random writes
3. 5mn sequential writes
4. TRIM (freeing up of cells) on 8 GB
5. 5mn sequential writes
6. Reset SSD (secure erase)
7. Filling the SSD
8. 5mn sequential writes
9. 20mn of random writes
10. TRIM (freeing up of cells) on 8 GB
11. 20mn of random writes
The periods of 5 and 20 minutes were fixed according to the speeds and overprovisioning capacities of 120/128 GB SSDs, so as not only to write to the overprovisioning capacity. The TRIM command was only used where indicated, with writes carried out continually otherwise without having recourse to TRIM, which is an extreme case. We didn’t, then, test the 512 GB versions of the M4 and Octane here but rather the 128 GB versions - the controllers behaviour is identical.

[ IOPS ]  [ MB/s ]

When we wrote sequentially to the SSD after it was full ("Full"), the performance of the M4 didn’t change at all. After the random write however, sequential write performance was down (“Used”). As the test continued to run performance levels returned to their initial level and freeing up 8 GB of flash cells with the TRIM command saw performance return to its initial level (“TRIM”).

The Octane 128 GB is in any case slower in sequential writes but once again performance didn’t suffer after the SSD was filled. After wear its performance was less affected than that of the M4 but it only recovered (almost) to its initial level after the use of TRIM whereas the M4 manages to do so without recourse to the TRIM command.

[ OCZ Octane ]  [ Crucial M4 ]

When it comes to random writes the Crucial M4 rapidly suffers after being filled to stabilise around 30 MB/s in place of 76 MB/s initially. Performance after wear is identical while freeing up 8 GB allows it to get back up to 60 MB/s and stabilize around 50 MB/s.

On the Octane, initial performance is lower, at around 24 MB/s. The test after filling the SSD shows that it stabilises around 10 MB/s with identical performance after wear. After TRIM, it returns to 24 MB/s and stabilises around 15 MB/s.

In the end, neither of the two SSDs does all that well in this extreme situation, with both suffering from a deterioration in performance. It isn’t too serious however and the M4 128 GB enjoys an advantage over the Octane.

Page 5
Energy consumption and Conclusion

Energy consumption
Finally we measured energy consumption with a clip ammeter. Consumption was taken at idle and during a sequential read and a sequential write, with this last task being the one that puts most demand on the SSD.

We have already noted several times that SSD consumption increases significantly with capacity. These SSDs are no exception to the rule with consumption levels of around 4.5 Watts during writes. Given that the Octane is faster with writes, it is in fact more efficient in this area. It is also more economical in reads but has higher energy consumption at idle.
With the Octane, OCZ has given itself a new lease of life in the world of SSDs. At the forefront of SSD innovation for the last four years via privileged partnerships with Indilinx and then SandForce, OCZ is now able to design its SSDs from the bottom up. Of course, in comparison to Crucial, Intel or Samsung, OCZ is still lacking a flash production facility and Hynix or Toshiba may, for example, soon find themselves being targeted.

It has to be hoped that OCZ will benefit from its new standing to avoid reproducing the mistakes of the past. We’re thinking here of the way the Vertex 2s were switched over to 25nm flash without any change in model number and even with lower capacities, or of the reliability issues discovered on the SandForce SF-2000 SSDs before the latest firmware updates. Of course, responsibility in this second case would seem to lie with SandForce, but as the assembler of the product, OCZ is answerable to the end user. We didn’t come across any such problems during our test with the OCZ Octane, but many consumers will have the company’s bad rep in mind when they come to buy their SSD.

All things considered, we can only commend OCZ’s metamorphosis, though the Octane struggles to stand out from the competition. In its 512 GB version it manages very high write performance, but this isn’t transposed to the other versions, which drop back to more standard speeds in comparison to the competition. With random accesses the Octane is down on the Crucial M4 (both read and write), so much so that application performance levels are slightly down. Note however that the difference isn’t really noticeable.

At the end of the day the OCZ Octane joins the pack of top SSDs that combine an SATA 6G controller and synchronous MLC memory, whether based on a Marvell (Corsair Performance Pro, Crucial M4, Intel SSD 511), Samsung (Samsung PM830) or SandForce SF-2000 controller (Corsair Force 3 GT, OCZ Vertex 3, Kingston HyperX). In practice these SSDs all offer very similar performance and warranties (3 years) and it’s therefore the price per GB and reports from other users that will influence you in your choice of which to go for. As the last out of the blocks, the Octane obviously has a bit of ground to make up, though even SSDs that are reputed to be reliable might happen to suffer from a bug...

Copyright © 1997-2014 BeHardware. All rights reserved.