Comments Locked

36 Comments

Back to Article

  • AFQ - Wednesday, November 28, 2012 - link

    Kristian, which software is this that you used to test the drives? and the first graph is of new RAID array. Second is without TRIM and third is with TRIM, right?
  • Kristian Vättö - Wednesday, November 28, 2012 - link

    I used Iometer for filling and torturing the drives and HD Tach for testing (it writes compressible sequential data across the whole drive). HD Tach is no longer being updated but you can still get it:

    http://downloads.guru3d.com/HdTach-3.0-download-20...

    I believe HD Tune has a similar graphing feature as HD Tach, so it might work as well.

    The first graph is after secure erase to give us clean-state performance. Second graph is without TRIM and the third one is with TRIM.
  • AFQ - Wednesday, November 28, 2012 - link

    Can you do a little guide about how to test the drive with IOMeter? Its like really advanced and technical.
  • vol7ron - Wednesday, November 28, 2012 - link

    You should ask these questions in the forums.
  • JonnyDough - Friday, November 30, 2012 - link

    Troll.
  • ney2x - Wednesday, November 28, 2012 - link

    First comment, first great news! Congrats Anandtech.com and Kristian Vättö (Hellhammer)
  • mooninite - Wednesday, November 28, 2012 - link

    Linux supports TRIM on the RAID layer on any chipset without the need for a hack or tweak.

    I have a RAID 0 on two X25-V drives on a old P45 chipset. I'd be SOL if I ran Windows.
  • Kristian Vättö - Wednesday, November 28, 2012 - link

    But what you have is a software RAID-0 array, correct? Software RAID-0 and TRIM has worked for ages but "real" RAID arrays created using Intel RAID ROM have not supported TRIM until now. I don't think the OS is the key here as long as it has proper driver support; the OROM didn't support TRIM on RAID-0 arrays before.
  • ender8282 - Wednesday, November 28, 2012 - link

    Is the Intel RAID truly 'hardware RAID' or is it just like the vast majority of other on-board RAID controllers the work is being doing by software running in the OS (maybe down at the driver level)?
  • Kristian Vättö - Wednesday, November 28, 2012 - link

    Intel RAID is often called "firmware" RAID because it's not truly hardware but it's not the same as software either. Software is strictly software and will only work with one OS because the RAID is loaded when the OS is. I believe firmware RAID is a combination of both because you need drivers for it but it works with any OS (even BIOS sees the drives as one volume).
  • ender8282 - Wednesday, November 28, 2012 - link

    OK that sounds a lot like what ever it was that I had on an old Abit motherboard. My experience with that setup was that unless the correct drivers were installed under windows, things didn't work right. Under Linux (mandrake I believe) it showed up as a bunch of different drives. Unless firmware raid has come a long way and supports more operating systems then just Windows, I fail to see how it is any better than software raid on Linux.

    Does software RAID under windows support trim?

    On a related note has AnandTech ever done any benchmarks between hardware, firmware, and software raid? I would be interested to read that article...
  • ender8282 - Wednesday, November 28, 2012 - link

    Probably a biased take but I found this [1] whitepaper gave a nice summary of the difference between hardware, firmware, and software.

    [1] http://www.adaptec.com/nr/rdonlyres/14b2fd84-f7a0-...
  • tecknurd - Wednesday, November 28, 2012 - link

    I disagree. The BIOS is firmware of the motherboard. Firmware is basically software for hardware that is loaded upon boot up and is not connected to the operating system in anyway. The BIOS includes RAID as software. When RAID it utilized from the motherboard, the software is loaded into memory. Mainly Windows will see this RAID software as any hardware RAID controller. Other operating system like Linux will not unless a userspace software is run first like dmraid.

    Stating Intel RAID as "firmware" RAID is a bunch of FUD. Intel RAID just does not have support to carry over the TRIM command compared to other modes of the Intel controller like AHCI.

    IMHO, Intel is all about pristine. The hack you explain is probably unsupported by Intel because is not stable and not reliable by Intel's standards.
  • extide - Wednesday, November 28, 2012 - link

    It is firmware RAID, and the software part is the driver. The RAID calculations are done in software on the main CPU, basically in the drivers. You would also need a matching driver in linux to use the 'firmware raid' as opposed to using completely software raid like dmraidc mdadm, etc.
  • tecknurd - Wednesday, November 28, 2012 - link

    Again it is not firmware RAID until you provide proof.
  • mooninite - Thursday, November 29, 2012 - link

    mdadm is used for Intel RAID as well as software Linux RAID. The Intel BIOS just writes out a special file format just like a file format for software Linux RAID. There is no decoder chip or firmware interface.
  • JonnyDough - Friday, November 30, 2012 - link

    I don't think you know what you are even saying. You must work for Intel and are trying to disprove this working fix because of "profit margins." The code was written by Intel and made to work by community software hax. Get over yourself.
  • mooninite - Thursday, November 29, 2012 - link

    Kristian,

    Intel RAID is *not* hardware. It is a software file format of the drive called IMSM. The OS has to do all the work to read and write to the drive. There is no hardware RAID chip involved.
  • CeriseCogburn - Saturday, December 29, 2012 - link

    You still are SOL mooninite, or you'd have Linux on a decent motherboard, and then you wouldn't need it anyway.

    Oh well that's how the tor valds.
  • Conficio - Wednesday, November 28, 2012 - link

    Anyone experience (or read about) how this works in a hackintosh?
  • vailr - Wednesday, November 28, 2012 - link

    There's also an updated motherboard bios for the Gigabyte GA-EP45-UD3R that includes the updated Intel RAID firmware 11.6.0.1702:
    http://forums.tweaktown.com/gigabyte/48085-gigabyt...
    So: updating most P45 chipset boards should also be a viable option.
    Also:t you're using an older 11.6.0.1030 RST driver.
    Latest version: 11.6.2.1002 WHQL
    http://www.station-drivers.com/page/intel%20raid.h...
  • Impulses - Wednesday, November 28, 2012 - link

    I'm very tempted to try this, now that I have two 128GB Samsung 830s, just not sure it's even worth the slight hassle... The instructions look simple enough, but what I'll gain is mainly the use of both drives as a single volume, since the increased sequential I/O isn't gonna matter much when every other drive on my system is a mechanical drive. I doubt any of my apps are really gonna see an usability improvement going from 500/350 sequential read/write max to 1000/700. :p
  • Sunburn74 - Wednesday, November 28, 2012 - link

    If you feel your ssd is fast enough then fine, but honestly there's a reason sad makers are still pushing speeds up as best they can. Heck its why sata3 is in the works. If ocz or corsair released a sad with sequentialsin the 1000s and random performance similar to the rest of the pack they'd be lauded as kings. Raid 0 accomplishes the same thing and now the one penalty has been made a non issue
  • Sabresiberian - Wednesday, November 28, 2012 - link

    I actually didn't know Intel had made their RAID chip TRIM friendly at all (and was wondering what was taking them so long, heh), so this article is very good news for me, even though I don't have a 6-series MB.

    Thanks Kristian for the great article covering a topic most power builders will be interested in.

    ;)
  • vol7ron - Wednesday, November 28, 2012 - link

    I'm not sure it's necessary to secure erase SSDs. At least, I didn't think it was in the past. Perhaps this would be something worth evaluating for the future.
  • Gigaplex - Wednesday, November 28, 2012 - link

    It's necessary to clean up LBAs to return back to stock performance when there's no TRIM support. It's not just a case of overwriting all the data on the drive like with mechanical drives (especially with SandForce compression/deduplication), there are special SSD tools that send a secure erase instruction to the drive.
  • extide - Wednesday, November 28, 2012 - link

    The secure erase always returns the SSD to a "like new clean slate" state. This is especially important in articles like this where you need to always start from a know control state that is always the same.
  • Sunburn74 - Wednesday, November 28, 2012 - link

    Secure erase wipes all data from the drive including the is. It was a tremendous pain in these if your using it to maintain performance. In addition you burn through write cycles quicker.
  • CeriseCogburn - Saturday, December 29, 2012 - link

    In spite of the hassle, first of all you're not "burning through" write cycles quickly, so there's no giant crisis "burning through them even quicker".

    You're crawling very slowly, nearly imperceptibly through write cycles.

    You will throw away the drives for being pathetically slow before write cycles are an issue, and that's what will burn you in the end.
  • Truegee1985 - Friday, November 30, 2012 - link

    I want to know if Intel's SRT version 11.6 trim support for Raid-0 on z77 chipset also work with Windows 8 Pro. I am currently using a z77 motherboard with an Ivy bridge cpu.
  • TheBeagle - Monday, December 3, 2012 - link

    While I certainly appreciate all the technical discussion in this and other similar articles, it would be a very great service to the enthusiast, as well as the average user, to simply and straightforwardly set forth what forms of RAID with TRIM can be used with SSDs, and any special circumstances where such a general statement of usability does not apply.

    There is a whole lot of mis-information circulating on the Internet about RAID and TRIM, and some simple guidelines would be of serious assistance to dispel the myths. Any such guidelines should include the minimum hardware and driver versions that need to be employed to obtain operational TRIM within a RAID array. Thanks. Best regards to all. TheBeagle
  • hechacker1 - Monday, December 10, 2012 - link

    I was updating my bios with intel's (at the time) new 11.5 rom, but on my x58 chipset.

    After I did that, I noticed that the SRT functionality was available to me. With one 11.5 version, the option was available, the next version, it was greyed out. I wonder if SRT is also just firmware/CPU based and really doesn't depend on the controller.
  • starwarsmike - Thursday, January 3, 2013 - link

    Hi firstly thanks for above info much appreciated....

    Can someone tell me if the above method to allow trim pass-through works with windows 8?

    Cheers Michael
  • Firebat5 - Sunday, December 8, 2013 - link

    Thanks for bringing this info to light. Wasn't sure if Fernando's forum was legit. With endorsement from you guys I went ahead and flashed a modified BIOS (thanks Fernando!) with the upgraded OPROM. Thanks guys!
  • i3v - Friday, June 10, 2016 - link

    I suspect, that the proposed method may not work, if SSD does not support DZAT (Deterministic Read Zero After Trim). This document provides some details, related to DZAT vs Deterministic Read After Trim (DRAT). Missing these two "features" does not actually mean missing TRIM.

    Here's a post about an SSD that only supports DRAT: http://forums.sandisk.com/t5/SanDisk-ReadyCache-SS...

    Another example: my "Kingston SHPM2280P2H/480G" seem to have neither DRAT nor DZAT, at least hdsentinel ( http://www.hdsentinel.com/faq.php#trim ) shows that "Deterministic read after Trim" is not supported. Thus, trimcheck-0.7-win64.exe shows that TRIM is off. Still, my Windows 10 correctly detects that this drive is an SSD (disk optimizer tool suggests to TRIM, not to defragment). And hdd sentinel shows that TRIM is supported and functioning (even though I do not understand how it checks this)
  • i3v - Friday, June 10, 2016 - link

    PS. It looks like the first link in my previous message is missing. Here it is: https://www.media-clone.net/v/vspfiles/downloads/D...

Log in

Don't have an account? Sign up now