Intel X25-M V2 (Postville) - BeHardware
>> SSD

Written by Marc Prieur

Published on July 31, 2009


Page 1


Intel made a lot of noise when it arrived on the SSD market in September 2008. With an ultra-high performance 10 channel controller, very effective wear levelling and limitation of writes on memory blocks, the Santa Clara giant had come out with an SSD with as yet unseen performance and out of the ordinary reliability with 100 GB/day writes for 5 years on MLC drives.

Unfortunately our enthusiasm was chilled due to a problem with stability of performance that showed itself in our first tests of the drive. After speaking of “expected behaviour”, Intel did correct the problem with a new firmware in April in response to criticism from tech media on the other side of the Atlantic.

With this problem resolved, the X25-M was once again an intriguing proposition though in terms of price per GB somewhat behind the recent competition from manufacturers using the Indilinx Barefoot controller (OCZ Vertex, GSkill Falcon, SuperTalent UltraDrive ME, Crucial M225, Corsair Extreme …) or Samsung S3C29RBB (Samsung PB22-J, OCZ Summit, Corsair Performance …). Initially planned for the last quarter of the year, the new X25-M is anticipated with a great deal of interest.
34 nm Flash chips
To start with, here’s how to distinguish the previous X25-Ms from these new ones, code name Postville :

- Old: SSDSA2MH080G101 and SSDSA2MH160G101
- New: SSDSA2MH080G2C1 and SSDSA2MH160G2C1

The main difference between these two revisions lies in the type of MCL NAND Flash used. Micron, Intel’s partner for everything Flash, is still the supplier but while the chips were previously manufactured at 50 nm, they now the new 34 nm process. This brings two advantages: the cost per GB is lower because of higher density, and the chips give better performance.

The old 160 GB X25-M used twenty 8 GB chips, while the new one uses 10 16 GB chips, which leaves the door open for a 320 GB version. This Flash memory is faster, Intel announcing read latencies at 65µs instead of 85µs and write latencies of 85µs instead of 115µs. In practice random reads of 4 KB blocks remains identical at 35000 IOPS, while for random writes of the same size we are up from 3300 to 6600 (80 GB) and 8600 (160 GB) IOPS. Sequential speeds are identical with reads at 250 MB/s and writes at 70 MB/s.

When you open the SSD you discover another difference: the memory cache is up from 16 to 32 MB. While previously Samsung SDRAM was used, now you’ll find it’s a Micron chip.

To give you an idea of the reduction in price over the last 10 months, you should know that the first X25-Ms were on sale for $595 and $945 for the 80 GB and 160 GB versions respectively. The new X25-Ms are at $225 and $440, a cut of 62% and 53%! Compared to the most recent prices for the X25-M V1s, the V2s are between 25% and 30% cheaper.

Page 2
Usure and the X25-M V2, the test

The usure of the X25-M in practice
Here is the usure of the X25-M with the new firmware. For a full recap of this phenomenon as well as the results obtained on the Vertex with the same protocol, read this article (pages 2 and 3).

Case 1: Sequential performance – Sequential access only

First of all, here are the sequential read and write speeds (in KB/s) when we carry out only this type of access (one read, one write, one read, one write, one read and a final write, each time on the whole of the SSD):

There isn’t a great variation, with read speeds around 214-209 MB/s and writes at 80 MB/s to 78 MB/s.

Case 2: Sequential performance – Random then sequential access

This time we started with a new drive (after HDD Erase) that is then “used” by writing by block of 4 KB randomly for 30 minutes. We carried out the same test as in case 1.

This time there was a significant fall in speeds with reads dropping to 193 MB/s. The fall is nevertheless slight and not as much as the fall recorded for the first generation X25-M. Moreover, after having carried out sequential writes the SSD returns to original levels of performance as of the second run.

For writes, we recorded 77 MB/s on the first run and then 78 MB/s for the next two, results that are therefore quite stable. With its original firmware the X25-M V1 dropped to 20 MB/s on this type of test!
Case 3: Random performance
Now here is an observation of the change in performance (in I/Os per second) during random writes of 4 KB files. We launch the test on a new SSD, 6 x 5 minutes, then a second time on an SSD whose cells have been previously filled sequentially.

On a new SSD, the X25-M V2 starts with a very high level of performance, more than 8000 I/Os per second as against 6500 for a V1. However it falls rapidly to 2000 I/Os before climbing back up slowly. On the V1, with the same test we finished the 6th run at 1262 I/Os; a clear improvement then.

On a used SSD, performances are of course lower with 5664 I/Os on the first run with the V2 as against 3022 for the V1. The least good result was on the 3rd run with 517 I/Os; the V1 scored 386 I/Os so once again the improvement is there for all to see.
The test
For this test, we compared the Intel X25-M V2 160 GB with different models:

- Intel X25-M V1 80 GB
- OCZ Core V2
- OCZ Apex
- OCZ Vertex 30 & 120 GB
- Samsung PM410
- Samsung PS410
- Samsung PB22-J 64 GB
- Samsung PB22-J 256 GB
- Mtron MOBI 3500

The OCZ Core V2 represents what is “best” with the JMicron JMF602/MLC combination, with the Apex a RAID version (two SSDs in one). The Vertex is based on the most recent Indilinx controller. The Samsung SLC and MLC SSDs are in a way the references for the previous generation, and the Mtron MOBI 3500 is an interesting alternative. The Samsung PB22-Js are the new generation of Samsung SSD, here in two capacities as their specs are different. For information we also include the performances of a VelociRaptor, a 3"1/2 Samsung SpinPoint F1 640 GB drive and a 2"1/2 Samsung SpinPoint M5 160 GB.

Different measurements were taken in the course of this comparison. First of all we were interested in the “synthetic” performances of these drives: average access time, sequential speeds and I/O in random and sequential access. Next, were more practical tests, first of all involving an applicative performance index based on PC Mark Vantage and then an evaluation of writing, reading with various groups of files These groups were composed of:

- A collection of large files: 6 files (on average 2.2 GB) totalling 13.2 GB
- Medium sized: 7.96 GB of 10,480 files (each averaging 796 KB)
- Small sized: 2.86 GB of 68,184 files (each averaging 44 KB)

The source or target of reading or writing on the drive was a RAID of three VelociRaptor 150 GB drives that replace the raid of two 150 GB Raptors used in previous tests. This type of measurement is worthwhile because, while the sequential speed gives us an idea of the performance when copying large files, things can be different with smaller ones.

The test machine was based on an X38 chipset mounted on an ASUSTeK P5E motherboard while Serial ATA ports were configured in the bios in AHCI (Advanced Host Controller Interface) so that NCQ could be used, all operating with Vista SP1.

We also added new practical tests so as to provide more useful data for those who want to work out if going for an SSD over their standard hard drive is worthwhile. For this we timed various operations on an other machine based on a P5QC, QX9770, GTX 280 and 2x2 GB of DDR2-1066.

Page 3
Synthetic performance

Synthetic performance
Mesured with h2benchw, SSD cache speed is really astonishing. At 0.12 ms, the X25-M V2 gets a very good scrore, the best of MLCs along with the Vertexes.

We continue with the speed, still with the help of h2benchw. In contrast to other benchmarking software such as HDTune or HDTach, h2benchw carries out a true sequential test because it reads or writes the entire drive, whereas the others jump between zones to reduce test time.

The X25-M V2 is very much at ease with read speeds of 214 MB/s, in line with the competition, but write speeds are not as good at “only” 80.5 MB/s. This is a bit better than the V1 but way behind the results obtained on some SSDs of the competition.

Now on to measurements of the number of 100% random accesses that can be supported by these storage systems, using IOMeter. We carry these out in small blocks of 4 KB and see how many the SSD can support each second, first using a type of sequential access and then random access. For this test we use a configuration file that carries out 10 tests for 10 minutes per test: 0, 25, 50, 75 then 100% random reads then 0, 25, 50, 75 and 100% random writes.

Sequential reads: the X25-M V2 leads with 14031 I/Os per second, ie over 54 MB /s. Random performance is lower as with all the SSDs, but by much less than with hard drives. The X25-M V2 therefore does better than the V1 but remains behind the Vertexes and the MOBI 3500.

In writes hard drives do better than in reads, the advantage being linked to cache size. For SSDs however, things are harder, especially as the test was carried out on a device on which the pages had already been filled previously with sequential writes.

In spite of everything the X25-M V2 does pretty well and is 67% better than a V1, with only the Vertexes ahead. Note that in the tests on the previous page that were made up of random writes only, the X25-M V2 did better than this.

Page 4
PC Mark Vantage

PC Mark Vantage
We now move on to less synthetic tests, starting with an index of hard drive performances in PC Mark Vantage. FutureMark reproduces a set of reading/writing tasks on the drive, namely a Vista start-up, loading of applications (Word, Photoshop, IE, Outlook), the manipulation of multimedia files (photo, video, music), games (loading of Alan Wake) and disk scan with Windows Defender.

With 27,347 points the X25-M V2 is in the lead, just in front of the… V1!

Page 5
Management of files

File copying
This brings us to file copying. We measure reading and writing of diverse groups of files. These groups were composed of:

- A collection of large files: 6 files (on average 2.2 GB) totalling 13.2 GB
- Medium sized: 7.96 GB of 10,480 files (each averaging 796 KB)
- Small sized: 2.86 GB of 68,184 files (each averaging 44 KB)

The source or target of reading or writing on the drive was a RAID of three VelociRaptor 150 GB drives.

Reading large files the X25-M V2 is very slightly down on the V1 but scores remain good, above all for large files. With smaller files, performances are comparable to a VelociRaptor.

Write speeds are the weak point on MLC Intel SSDs. Happily although the speed with large files is slow, the V2 does well on medium-sized files. What’s more, with small files the gain you get with the V2 puts it out in front of all other SSDs and just behind the SpinPoint F1, which is particularly fast here.

Page 6
Practical tests

Practical tests
We have at last decided to add other practical tests to our SSD tests, in particular so as to provide more useful data for those who want to work out if going for an SSD over their standard hard drive is worthwhile. For this we timed various operations on an other machine based on a P5QC, QX9770, GTX 280 and 2x2 GB of DDR2-1066.

- Windows Vista start-up
We measure the time needed to start up Windows Vista freshly installed with the drivers. The measurement is from running the bios (disappearance of P5QC logo) to full Windows desktop display with the cursor not showing the egg times.

- Installation of Service Pack 1
Here, the time needed to install Service Pack 1, the installation file itself being situated on the SSD.

- Start up of Windows Vista SP1+Kaspersky+Word+Excel+Outlook+Photoshop
After having installed Kapersky Antivirus 2009, the Office suite and Photoshop CS4, we put shortcuts to Word, Excel, Outlook and Photoshop in the Startup directory of the start menu. The time measured is from running the bios to the end of Photoshop CS4 launch.

- Loading of “Train” level in Crysis Warhead
After having installed and patched Crysis Warhead we launch it with the –DEVMODE option and load the train level from the consol with the “map train” command.

We weren’t able to carry out the tests on the 256 GB version of the Samsung PB22-J, as we no longer had this SSD in the lab.

On start up of a “new” Vista, the X25-M V2 is the fastest, even though you have to say the differences between the various SSDs is small.

The installation of SP1 is to the advantage of the X25-M V1 and V2, with the Vertexes also doing well.

While there were no major differences with just the start-up of Vista, it was different when we launched very heavy multi applications. The X25-M V2 is in the leading group along with the V1, the OCZ Vertexes, the Mtron MOBI 3500, and the Samsung PS410 is out in front.

As you can see, moving from an HDD to an SSD doesn’t necessarily change much in terms of loading of games. Of course, during loading there are read operations that are carried out on the storage device but also a lot of the time is linked to the creation of the environment by the processor from this data – this time can’t be cut down. Therefore if there is a gain it is slight and the X25-M is well placed.

Note that in the end in these practical tests the X25-M V2 160 GB gives only very slightly better performance than the X25-M V1 80 GB, this in spite of any more significant differences that may be present in synthetic tests.

Page 7
Power consumption

Power consumption
All the SSDs tested here are completely silent. We concentrated therefore on measuring consumption at rest and reading/writing with IOMeter. Here are the results obtained:

Levels of consumption are fine and you’ll see that although the X25-M V2 160 GB is no better in idle than the X25-M V1 80 GB, in load this falls from 2 to 1.1W!

Page 8

Although there are clear improvements with the X25-M performances in synthetic tests, particularly for random writes, the difference in practice between the two versions of the Intel MLC SSD is more than slight. Some will regret that the major weak point of the series, namely sequential write speeds, has not been corrected, but you have to say this is very much secondary for an SSD given what they are used for.

Overall the X25-M V2 gives, like the V1, top drawer performance, making it a choice option for those looking for, well, a top drawer SSD. The cut in costs makes the pricing of the V2 very competitive in terms of price per GB and it has the advantage of giving very stable performances over time without having to resort to TRIM as is the case with Indilinx SSDs. In any case the 34 nm X25-M should, like SSDs based on an Indilinx Barefoot, benefit from a new firmware that supports TRIM native to Windows 7 as of the official release of Microsoft’s new OS. To recap, this command integrated into the OS allows the SSD to know which Flash pages are no longer used so as to maintain optimal performance levels. It will be a pity if, as is rumoured, the old 50 nm X25-Ms do not benefit from this update, which would seem, nevertheless, to be technically possible. Such a choice would certainly not meet with our approval if confirmed!

A last word on the reviewed firmware. It had a bug that could be annoying: if you used a password in the bios and you modified it or disactivated it later, the X25-M V2 became unusable and the data innaccessible. Hopefully Intel has now fixed the firmware and X25-Ms V2 shipped now won't come with this issue.

Copyright © 1997-2014 BeHardware. All rights reserved.