Home  |  News  |  Reviews  | About Search :  HardWare.fr 

MiscellaneousStorageGraphics CardsMotherboardsProcessors
Advertise on BeHardware.com
Review index:
OCZ Octane 512 GB and Indilinx Everest vs Crucial M4 512 GB
by Marc Prieur
Published on February 23, 2012

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.

<< Previous page
Practical performance – File copying and applications

Page index
1 | 2 | 3 | 4 | 5
Next page >>
Energy consumption and Conclusion  

Copyright © 1997- Hardware.fr SARL. All rights reserved.
Read our privacy guidelines.