OCZ Vertex SATA 2.0 60GB SSD

[curaga]

The aligning to erase blocks works with SD cards at least.

OTOH I recently read an article about how Windows' 1mb aligning was just plain wrong, but IIRC the article was about EFI/GUID, not SSD.

[deanjo]

This review is completely flawed, unless I missed something.

The partition is not aligned, therefore you get degraded performance. Windows Vista and 7 aligns the partition automatically and correctly (1MB offset). See this thread: http://www.ocztechnologyforum.com/fo...ad.php?t=54379

You'll want to use the right io scheduler (deadline seems to be ok), the right ext4 mount options, and correct alignment.

Interesting, I wonder if the same could be applied to flash drives.

[fackamato]

This review is completely flawed, unless I missed something.

The partition is not aligned, therefore you get degraded performance. Windows Vista and 7 aligns the partition automatically and correctly (1MB offset). See this thread: http://www.ocztechnologyforum.com/fo...ad.php?t=54379

You'll want to use the right io scheduler (deadline seems to be ok), the right ext4 mount options, and correct alignment.

[tomm3h]

Ah, well thanks for clearing that up.

[Ex-Cyber]

Shorter version: Ext4 supports TRIM, but it's disabled because they can't test it (because no drives actually support it).

Longer version (from this interview):

Ext4 has support for the ATA TRIM command, which allows filesystems to inform SSD?s that blocks have been deleted and do not need to be taken into account by the SSD?s garbage collection and wear-leveling algorithms. Unfortunately the ATA TRIM command hasn?t been finalized yet, and so (as of today) there are no drives, including Intel?s SSD?s that actually support the ATA TRIM command; and for this reason Linux?s block device layer does not currently issue the ATA TRIM command, since there haven?t been any devices to test the command. So at the moment, ext4 informs the block layer that blocks that belong to deleted files can be discard, so once TRIM-capable SSD?s become available, and the Linux block layer actually sends the TRIM command to the hard drives, everything will be all set to go.

[tomm3h]

Here it is the anand article, I think the best they wrote.

http://www.anandtech.com/storage/showdoc.aspx?i=3531

One of the biggest problems of today SSD's is the performance degradation, the more you store data, the drive get significant performance loss, up to 11% of a erased one. Intel made a pre-boot software that complete restore theyr SSD's, cleaning all data of course.

Thanks for posting the link.

With regards to the TRIM command, I'm not even sure if Linux supports it yet? I can't see anything on Google... It will be a shame if Windows 7 ships with support for it before Linux.

[dm0rais]

For now, your latter assumption is correct. It's not just the controller, but the firmware used. OCZ were originally going to ship the Vertex with more performance, but its firmware was found to be stalling repeatedly (Anand wrote a very good article on this, if you can find it.)

@Michael: if you're intending on using this drive in one of your main machines (I would) then it would sure be a nice idea to see the same tests run after ~3 months of solid use on the disk, just to ensure that there isn't a disproportionate level of performance loss over time.

Same tests, same OS, just comparing the results for the Vertex now, to the results of the Vertex then. That is, if you're allowed to keep it...?

Here it is the anand article, I think the best they wrote.

http://www.anandtech.com/storage/showdoc.aspx?i=3531

One of the biggest problems of today SSD's is the performance degradation, the more you store data, the drive get significant performance loss, up to 11% of a erased one. Intel made a pre-boot software that complete restore theyr SSD's, cleaning all data of course.

[tomm3h]

So, when I buy my next SSD.. Which controller do I want to look for? Indilinx something or other? ..or should I just stick to OCZ Vertex and Intel X-25 for current considerations?

For now, your latter assumption is correct. It's not just the controller, but the firmware used. OCZ were originally going to ship the Vertex with more performance, but its firmware was found to be stalling repeatedly (Anand wrote a very good article on this, if you can find it.)

@Michael: if you're intending on using this drive in one of your main machines (I would) then it would sure be a nice idea to see the same tests run after ~3 months of solid use on the disk, just to ensure that there isn't a disproportionate level of performance loss over time.

Same tests, same OS, just comparing the results for the Vertex now, to the results of the Vertex then. That is, if you're allowed to keep it...?

[ethana2]

One has to wonder why SSD's carry a shorter warranty period then traditional harddrives if they are indeed more reliable.

Indeed.

So, when I buy my next SSD.. Which controller do I want to look for? Indilinx something or other? ..or should I just stick to OCZ Vertex and Intel X-25 for current considerations?

[deanjo]

or are looking for the greater reliability that is offered by solid state drives.

That has yet to actually be field proven. One has to wonder why SSD's carry a shorter warranty period then traditional harddrives if they are indeed more reliable.