Intel X25-E review

[Denithor]

Review at bit-tech.net

Blazing fast but they had this unsubstantiated comment at the end of their review that has me scratching my head.

Unlike with mechanical or MLC drives where data can be stored in multiple states, the SLC memory only modulates between written and unwritten. This means that once the drive has written to each cell, rewriting to them means the drive must first set the cell from written, to unwritten and then back to written in accordance with the new data, doubling the write times the second time the drive needs to write to that particular cell.

While fully formatting the drive resets all the cells to unwritten and brings back that awesome write performance evident in our graphs, it?s hardly reasonable for end users to have to format their drive every time it fills up and although we realise this is a very specialised enterprise level drive, it?s still a disappointing flaw that?s unfortunately part of using SLC memory in the first place.

Have any of you heard of this issue with SLC drives? They don't offer any evidence in the report to verify in any way this rather strange claim of reduced performance after the first "fresh" write to an unwritten sector on an SLC drive.

 

[Viper GTS]

Yes, Intel even documents this behavior. It takes the drive a while to reach a steady state performance, & to get accurate benchmark numbers you must "season" the drive beforehand.

FWIW Intel claims that the numbers they publish are the steady state numbers, not the inflated fresh-drive numbers.

Also - Their belief that MLC does not suffer the same problem is incorrect. The exact same effect occurs on the X25-M drives, and again is fully documented by Intel.

Viper GTS

 

[IntelUser2000]

Yes, Intel even documents this behavior. It takes the drive a while to reach a steady state performance, & to get accurate benchmark numbers you must "season" the drive beforehand.

The reason for the different performance values ARE BECAUSE THE CONTROLLER IS SO GOOD!!

Look here:

1st one, Oct 23, 2008
http://img177.imageshack.us/my...mage=hdtunex25may2.jpg]

[/URL]

2nd one November 4, 2008:
http://img101.imageshack.us/my...age=hdtunex25m2no1.jpg]

[/URL]

Test I did December 10, 2008:
http://img101.imageshack.us/my...dtunex25m3laterhe0.jpg]

[/URL] As you can see on the last pic, the dip shown at 20% disappeared after prolonged usage.

No review will show this.

Explanation about the dynamically optimizing controller:

http://hothardware.com/Article...l-Ups-The-Ante/?page=2

"Finally, and most importantly, Intel's flash memory controller offers dynamic workload adaptation such that the drive will actually adjust on the fly to specific workload conditions"

This is one of the reasons I believe the results for the X25-M varies quite a bit between multiple review sites. I'd assume the dynamically optimizing controller would work in conjunction with the wear levelling and it would take writing to the ENTIRE drive couple of times to get the usage patterns. Look how long my drive took.