Support and Q&A for Solid-State Drives

New windows features with SSD

nomad27 — Tue, 05 May 2009 11:26:39 GMT

Glad to hear you are taking new technologies into engineering considerations in W7.

Can you tell me if work is done to take into account working with non-volatile memory(memory that keeps the data even when power is turned off) such as MRAM?

The possibilities for new features are endless such as turning off the pc completely and instantly resuming without any data loss.

Windows will have to adapt to the new memory quickly and for example differentiate between restart shut-down and non-restart shut-down.

re: Support and Q&A for Solid-State Drives

FlemmingRiis — Tue, 05 May 2009 11:32:15 GMT

Yet another reason to enbrace Windows 7

Engineering Windows 7 blogged about SSD support in Windows 7

Manuel trackback

http://it-experts.dk/blogs/flemmingriis/archive/2009/05/05/windows-7-forbedringer-mht-ssd-diske.aspx

re: Support and Q&A for Solid-State Drives

Maciej Rutkowski — Tue, 05 May 2009 13:55:22 GMT

It strikes me as amazing to have that much knowledge under your fingertips thanks to telemetry of potentially millions of computers.

Just as Google with their hard drive use paper ( http://www.engadget.com/2007/02/18/massive-google-hard-drive-survey-turns-up-very-interesting-thing/ ), you could post some performance and reliability results based on your data.

A massive and very real-world benchmark would come out of that actually.

Good job MS. Been an "early" ;) RC adopter for two days now and the only problem I see is sometimes poor Aero performance of my AMD690G on-board graphics (Radeon x1200 series). Other than that it's snappy, beautiful and extremely usable and polished.

re: Support and Q&A for Solid-State Drives

someone — Tue, 05 May 2009 14:17:41 GMT

Will Windows reduce the number of times its background apps and third party apps (anti virus updaters) write data frequently to an SSD?

Also, will Encrypting File System use have any impact on SSDs?

re: Support and Q&A for Solid-State Drives

krish4u — Tue, 05 May 2009 14:37:54 GMT

Statement:

"Will Windows 7 support Trim?"

Yes.

So, what is the status with the Windows 7 release candidate? Is Trim already supported? How do I find/check this out?

Great job with the SSD support. I have one and Win7 RC literally flies .... :)

-krish

RC download experience

tom5 — Tue, 05 May 2009 14:55:20 GMT

A big thank you to MS for providing good download experience of the RC - you guys learned a lot from the public beta :)

re: Support and Q&A for Solid-State Drives

manicmarc — Tue, 05 May 2009 15:26:09 GMT

Where does exFAT come into this? If at all? Will Windows 7 Netbooks be formatted as exFAT by default?

re: Support and Q&A for Solid-State Drives

Nehemoth — Tue, 05 May 2009 16:08:16 GMT

Really, really excelent articules, one of the best That I ever read here or even anywhere concerning SSD performance.

Thank you

re: Support and Q&A for Solid-State Drives

tgrand — Tue, 05 May 2009 17:21:00 GMT

Everything sounds really great here except for fragmentation. Your solution is to just allow filesystem fragmentation to run unchecked and rely on the SSD to power through it? I thought trim might help enable a better defrag approach.

I'm curious about whether your measurements of performance degradation also took into account filesystem fragmentation. As you know, it's very common for some files to reach thousands and even tens of thousands of fragments, and free space can become extremely fragmented on highly utilized drives. I question whether SSD performance is sufficient to counter this using a brute force approach.

I wonder if you decided that it would be better to defer this task to third parties for now, and look at solving it in a future version of Windows. Or, if you genuinely believe this approach offers the right tradeoff of performance vs. wear long-term.

re: Support and Q&A for Solid-State Drives

sroussey — Tue, 05 May 2009 18:54:56 GMT

@tgrand: "As you know, it's very common for some files to reach thousands and even tens of thousands of fragments, and free space can become extremely fragmented on highly utilized drives."

They don't directly address it, other than some SSD drives have 50% more space than advertised to the system (most have only about 10%). But for anyone looking at a SSD drive, you should never fill them up. That will dilute wear leveling, and many performance benefits. Filled drives are the typical reason for such highly fragmented files.

That said, if the Win7 disk defragger had a SSD type mode that only defragged the rare cases (leaving the rest well enough alone), that would be a nice option.

I found that I was going to use about 45G for a system drive, figured on doubling that to get the wear leveling and opportunistic flash cell writing which had me looking for a 90G drive. The closest thing was a 120G vertex. It would have fit in a 60G, but I know how the engineering works, and would never have so little 'free' space. Unlike a HDD, that free space actually gets used.

re: Support and Q&A for Solid-State Drives

bananaman — Tue, 05 May 2009 20:03:05 GMT

If SSDs are attached to a hardware RAID controller (which typically identify their arrays using their own manufacturer's name) will Windows 7 know that the array is composed of SSDs and use the trim function?

Will there be a way to manually tell Windows that this array (or single device) is/are SSD's, this one is magnetic platters, and so on?

re: Support and Q&A for Solid-State Drives

swaroopk — Wed, 06 May 2009 03:54:15 GMT

@Maciej Rutkowski :

A massive and very real-world benchmark would come out of that actually.

Benchmarks based on real world traces are immensely valuable. Some of the traces referenced in the above USENIX paper as well as some other traces from production servers referenced in this IISWC paper have been made publicly available by Microsoft via SNIA to enable research by academia and others. Of course all personally identifiable information (PII) has been removed from these traces. Note that these are server traces and not client traces. System traces can be captured by using the built in ETW functionality in Windows (client as well as server) and visualized/analyzed using the Windows Performance Tools Kit.

@bananaman:

Will there be a way to manually tell Windows that this array (or single device) is/are SSD's, this one is magnetic platters, and so on?

Typically the disks behind a RAID controller are managed by the RAID controller and presented to the operating system as one or more units (disks) of storage space. If the RAID controller reports the rotational speed as zero for the units of storage (disks) that it presents to Windows 7 then it will treat that unit of storage as an SSD.

- SwaroopK (Windows Team - Microsoft)

re: Support and Q&A for Solid-State Drives

FrankyN — Wed, 06 May 2009 04:19:40 GMT

Not very long ago i installed WinXP on a usb stick after some heavy tweaking, and is running fine. And faster then on the internal SSD drive.

I know, not supported nor recommended. But the technology is advanceing, and therefore a possible option. The are getting bigger and faster.

Would it be possible for Win7 to install it on a memory stick? If not, a future option?

It is not the same as a SSD drive, but still Solid State.

Awesome article!

mbampo — Wed, 06 May 2009 05:49:43 GMT

The best so far in the blog. Thanks guys!

re: Support and Q&A for Solid-State Drives

dcuccia — Wed, 06 May 2009 10:20:55 GMT

Sorry if I missed it - will FAT be allowed on SSDs? Will defrag still be turned off in this case? If so, isn't this adding extra risk in a recovery scenario? Thanks, David

re: Support and Q&A for Solid-State Drives

Chris_7 — Wed, 06 May 2009 10:47:14 GMT

I've just signed in to leave a comment and ask a question ;)

In this article you said :

"Will Windows 7 support Trim?

Yes. See the above section for details"

But my question is :

IS Windows 7 supporting Trim right now (RC1)or TRIM is going to be supported by Windows 7 at the final release or in a Service Pack ?

Thanks a lot ;)

re: Support and Q&A for Solid-State Drives

kamil.nowicki — Wed, 06 May 2009 15:20:52 GMT

@SwaroopK

"If the RAID controller reports the rotational speed as zero for the units of storage (disks) that it presents to Windows 7 then it will treat that unit of storage as an SSD."

...and if controller can actually pass TRIM to the drives. to me that is too much "if's" to believe that TRIM on RAID will work from day 1.

Does CR1 have Trim?

DarinMS — Wed, 06 May 2009 21:30:21 GMT

I have to echo (and expand) Chris_7's question.

Does Win7 RC1 have Trim?

If Win7 currently has Trim (or when it is added) what 3rd party work needs to be done for the end user to actually utilize Trim? (i.e. SSD firmware flash to accessp Trim command, support for Trim in Motherboard chipset drivers and so on)

Thanks for the blog entry, I've been looking for information regarding Win7 and Trim (just bought 3 OCZ Vertex SSDs for a couple new builds I'm working on)

Does RC1 have Trim

DarinMS — Wed, 06 May 2009 21:37:35 GMT

Oh, and if Win 7 RC1 doesn't currently have Trim, when is it expected to be added?

re: Support and Q&A for Solid-State Drives

craigbarkhouse — Wed, 06 May 2009 22:41:08 GMT

@Chris_7, DarinMS:

Yes, Trim is already in the Win7 RC.

Trim is enabled by default but can be turned off. You can use the "fsutil behavior query|set DisableDeleteNotify" command to query or set Trim.

re: Support and Q&A for Solid-State Drives

marcinw — Wed, 06 May 2009 23:58:41 GMT

Very funny thing. I had time to look into RC version and it's even worse than beta. We have issues from 7000 build and more. It shows, how Microsoft is treating own customers. You can delete my post, but facts will be facts (see below).

First example:

1. set UAC level to highest one

2. go into Computer Management

3. set Startup Type for Application Information service to Disabled

4. restart system

5. many system actions (requiring UAC prompts) will simply not work, you can't easy fix it

I understand, that it will need access to computer. But selling system to customers with such functionality "by design" won't be too honest and professional.

Second example: like zdnet.com notified, system by default hides files extensions. How many non-technical users will understand, that file "document.txt" in Explorer is "document.txt.exe" ?

Third: http://www.pretentiousname.com/misc/win7_uac_whitelist2.html (I haven't checked, but I believe, that it can be the truth)

Please note, that I haven't went into deeper things. For example why system doesn't allow for formatting me partition in exFat ? Why Explorer still can't enter directory, which is "link" ? Why icon for Network Sharing and Center doesn't display animation ? why clear type tuner doesn't have maximize button ? why after enabling displaying hidden files I have two files on desktop ? etc. etc.

I have feeling, that nobody (I repeat nobody) is controlling it. There is no one clear consistence vision (what to do with this architecture) and we have mix of everything with everything. This is one big mess.

can we override the default no-defrag?

Surt — Thu, 07 May 2009 00:05:47 GMT

I have the occasional situation where it would still be valuable to defragment, can we override the default policy and defrag our SSDs if we want to?

to marcinw

nomad27 — Thu, 07 May 2009 11:55:50 GMT

To marcinw

What are you saying?

That by doing these 2 actions, that no customer will or should ever do.. you got a bug?

So.. kudos to you. You're MS #1 beta tester - you found a bug in the RC.

And with that you conclude that "It shows, how Microsoft is treating own customers"?!

Go play programmer in linux or something.

re: Support and Q&A for Solid-State Drives

Igor Leyko — Thu, 07 May 2009 14:45:42 GMT

"Pagefile.sys read sizes are typically quite small, with 67% less than or equal to 4 KB" - how can pagefile read be less than 4 KB? This is the size of memory page and memory manager operates with pages not bytes.

And thanks a lot for this statistic, I've looked for it for a long time.

Igor P. Leyko

MVP System & Performance

re: Trim()

krish4u — Thu, 07 May 2009 16:53:43 GMT

@craigbarkhouse :

fsutil behavior query DisableDeleteNotify

returns a 0, which I take to mean to that Trim() is enabled?

The SSD I have is flashed to a firmware revision that supposedly implements Trim(), so the questions I have are:

- Is Win7 RC1 actually using the Trim() command in my case?

- If an SSD does not implement Trim(), would the above command return a 1?

Thanks,

-krish

re: Support and Q&A for Solid-State Drives

Wolf-Tech — Thu, 07 May 2009 18:48:58 GMT

I'm running Windows 7 RC 64 bit right now and so far I find disabling the Superfetch and the Prefetcher and Readyboot increases my boot up times. Also Once I disable Prefetecher I dont have that wasted 600 megs of memory or so for that stupid readyboot feature. My Windows 7 now boots under 6 sec to desktop only using 320 megs of ram.

re: Support and Q&A for Solid-State Drives

marcinw — Thu, 07 May 2009 19:39:32 GMT

@nomad27,

In earlier Windows versions disabling some services was blocked, in Seven "Application Information" is not blocked. You should do steps described by me and you will see, that normal user will need reinstalling system to get it working. Is this really hard to understand ? Authors of various malware are only waiting for such occasions...

This blog is read by many Microsoft employees (including Steven). Info about this issue in 7000 was written by me long time ago. And nothing... Many people want returning some things (animated icon with network activity is example). And nothing... Sorry, but this system is created for users. This is not art for art.

Microsoft fans were screaming "wow" in 2007 and we know results. Maybe it's time to stop screaming "excellent" in 2009 and start thinking, how to better address customer needs ? Please note - after first "excellent" opinions in more and more sites you can read opinion, that Seven is not so good (at least: it's not revolution and doesn't have killer features)

re: Support and Q&A for Solid-State Drives

craigbarkhouse — Thu, 07 May 2009 23:27:13 GMT

@krish4u:

Correct, if fsutil reports that "DisableDeleteNotify" is 0, then Trim is enabled. (The feature is sometimes referred to using different names: Trim == Delete Notification == Unused Clusters Hint.) The setting is written in terms of disabling something because we like to use values of 0 for defaults.

Have Trim enabled according to this setting, which you do, means that the filesystem will send Trim commands down the storage stack. The filesystem doesn't actually know whether this command will be supported or not at a lower level. When the disk driver receives the command, it will either act on it or ignore it. If you know for sure that your storage devices don't support Trim, you could go ahead and disable Trim (enable DisableDeleteNotify) so the filesystem won't bother to send down these notifications. However sending down the notifications is pretty lightweight and I haven't seen any performance improvement by disabling them, so I don't recommend disabling this setting. If you have an SSD which does support Trim, then you definitely don't want to disable it, because there are some performance gains to be had for leaving the setting in its default form.

re: Support and Q&A for Solid-State Drives

JamesD0109 — Thu, 07 May 2009 23:30:27 GMT

I'm intending to install the RC on a RAID0 made of two SSDs. Will the installer be able to detect that the drives are SSDs automatically? Is there a way to find out if it has? If not, is it possible to force it to detect a drive as SSD?

Slightly off-topic - on systems with SSDs and a lot of RAM (12gb, in my case) is there any point in having a page file above 4gb (say) if I'm not bothered about full memory dumps? SSDs generally being smaller than normal drives, I'd rather not give up a full 12gb to it!

re:re: trim enabled in win7 RC?

m.oreilly — Thu, 07 May 2009 23:30:38 GMT

indeed, is there any info as to an official response? apparently the "fsutil behavior query set DisableDeleteNotify" also gives the same values under the 7000 beta release...

re: Support and Q&A for Solid-State Drives

wtroost — Fri, 08 May 2009 01:34:27 GMT

Good read, especially the tip about putting a pagefile on the SSD. I would've expected the pagefile to be bad for the SSD because it would wear out the flash faster!

re: Support and Q&A for Solid-State Drives

craigbarkhouse — Fri, 08 May 2009 03:42:13 GMT

@m.oreilly:

Yes, fsutil shows the same value in the Beta and the RC, because Trim is supported and enabled by default in both the Beta and the RC. ;)

I should add that very few (if any) SSD drives in the marketplace today actually support Trim. Most of the ones that do are next-generation prototypes. But when they do become available, Windows 7 will take advantage.

- Craig (NTFS team)

re: Support and Q&A for Solid-State Drives

locolorenzo — Fri, 08 May 2009 06:11:16 GMT

I plopped a SSD into build 7100 RC and as stated it works like a charm.

There are a lot of really great things happening with this Operating System, I have people who come to me for their computer needs testing the RC...all of them will be purchasing the Final Release.

Thnx

re: Support and Q&A for Solid-State Drives

anonymuos — Fri, 08 May 2009 08:19:38 GMT

All "Certified for Windows 7" logoed drives will support trim right?

re: Trim()

krish4u — Fri, 08 May 2009 18:01:29 GMT

@Craig (NTFS team):

<quote>

</quote>

So do the Microsoft IDE Controller disk drivers support it? If you have a 3rd party disk driver provider (i.e. non-Microsoft), then I guess you rely on them implementing the Trim() functionality?

Thanks for your quick responses!

-krish

thanks craig

m.oreilly — Sat, 09 May 2009 01:34:23 GMT

thanks for the clarification. is this the final form of trim to be implemented in win7 (barring any future rtm update/sp)? i am using vertex drives which have, at this point, a somewhat proprietary, functioning trim fw, though we are expecting an ms compatible version in about a weeks time. can i assume you are "good to go" re win7 trim, and that drive controller manufacturers will have what they need?

re: Support and Q&A for Solid-State Drives

henriqueG — Sat, 09 May 2009 02:08:44 GMT

I'm using the Windows 7 RC and I love it. I have only a suggestion for you. To unpin some icon from the taskbar make something like drag-n-drop, to put drag the icon to the taskbar (like Win7 RC) and to unpin drag the icon to the desktop, disappearing with the icon

;)

Thanks...

re: Support and Q&A for Solid-State Drives

robmuld — Sat, 09 May 2009 02:40:25 GMT

PLEASE PLEASE PUT THE CLASSIC START MENU OPTION BACK IN WIN 7!!!!

re: Support and Q&A for Solid-State Drives

me&er — Sat, 09 May 2009 12:42:42 GMT

<<"I should add that very few (if any) SSD drives in the marketplace today actually support Trim. Most of the ones that do are next-generation prototypes. But when they do become available, Windows 7 will take advantage.

- Craig (NTFS team)">>

Thanks for your input so far Craig.. much appreciated.

While the bigger folk sort out SSD 'logo' reqs etc.. any 'issues' with a little fella like me trying to disable Win 7(RC) native TRIM and using a brute force propriety trimming 'tool' which would work really well for me on a customised 'schedule' basis. I intend to use SATA Native IDE mode and default Microsoft ATA/ATAPI device drivers with a combined MLC/SLC implementation.

By 'issues' I refer to legal/propriety as well as technical.. my assumption being that Procmon can keep me reasonably well informed of what's going on 'technically' between the OS and my SSDs.

Regards,

[OT] USERS folder shared with full access by default?

tom5 — Sat, 09 May 2009 23:34:00 GMT

Hi,

Just installed the RC and there's one thing strange to me: sharing is enabled by default for the USERS folder (not the public folder) and from my other computer I can access everything in this folder and perform operations (read/copy/delete). It was a clean Win7 install and of course I haven't changed the sharing policy for myself.

I now disabled the sharing and now it is as it should be.

re: Support and Q&A for Solid-State Drives

lukechip — Tue, 12 May 2009 04:30:15 GMT

At what level does Win 7 disable defragmentation, Superfetch, ReadyBoost etc for SSDs ? I have an SSD as the system drive in my Win 7 RC machine, and the defrag GUI shows scheduled defrag as turned on, and in the schedule, select disks includes the SSD. Is defrag for SSDs disabled at some lower level that the GUI does not show ?

I manually turned off scheduled defrag, suspecting that my SSD was not be handled correctly by the Operating System.

When I run fsutil behavior query DisableDeleteNotify, I get a 0. Does this mean that my SSD is being correctly recognised as an SSD ?

I do notice that on the ReadyBoot tab for an SD card I put into my machine, it says "This device cannot be used for ReadyBoost. ReadyBoot is not enabled because the system disk is fast enough...."

Finally, is there a way for me to determine what rotational speed my SSD is reporting to the Operating System ? fsutil ? wmi call ?

"NAND only SSD" Alternatives Exist and Offer Substantial Benefits for Targeted Applications

DDRdriveLLC — Tue, 12 May 2009 19:59:17 GMT

It is important to note, for IOPS intensive enterprise storage, there are alternatives to a "NAND only SSD". One alternative, the DDRdrive X1, elegantly avoids all of the above mentioned limitations of Flash by using DRAM for IO reads/writes and Flash solely for backup/restore.

1) No LBA remapping, thus no wear leveling overhead.

2) Deterministic performance, no pauses or stuttering when dealing with writes - ever.

3) No OS alignment issues, i.e. no performance penalty on Windows XP.

4) Unlimited reads/writes - no fundamental wear mechanism until drive performs a backup/restore.

5) Defragmentation can be turned on, with resulting performance increases.

The above benefits do come with the both capacity and price tradeoffs. Certain applications, for example databases, can be architected to easily work within these constraints.

The drive for speed,

Christopher George

Founder/CTO

DDRdrive LLC

www.ddrdrive.com

re: Support and Q&A for Solid-State Drives

craigbarkhouse — Tue, 12 May 2009 22:37:24 GMT

@krish4u:

I'm not an expert on our storage drivers (I deal mainly at the file system level), but it appears that our ATA port driver (ataport) does implement trim support. This means that SSD drives which present themselves as ATA drives (which I think most if not all do), can support trim provided the drive itself also supports trim. Non-ATA devices -- including USB drives and SCSI drives -- don't yet have the ability to support trim, since our other port drivers don't implement trim. This may change as the market evolves. I don't know if any 3rd-party storage drivers implement trim as of yet, but yes, they would have to implement it for it to work.

@m.oreilly:

As far as Win7 RTM goes, trim is in its final form. Of course it could evolve in service packs, etc., as the market demands. It's a pretty new market. Drive manufacturers know what they need to implement on their end. Some have provided prototypes to us for testing.

@me&er:

I doubt there are any legal issues with you sending down trim commands yourself, but it sounds like an awful lot of work to me. Firstly I'm skeptical that you can even get the proper information you need. I don't think you can infer from ProcMon output alone what clusters are in use and what aren't. The file system knows this; what makes you think you can do better? If you get this wrong, you can end up corrupting the volume. Secondly, I'm not sure what the benefit of your approach would be even if you could get it right. By having the file system send down trim commands when appropriate, you enable the drive to immediately benefit from this information. There's very little overhead to these commands. Contrast this with defragging, where if you were constantly defragging everything the cost would outweight the benefits. I strongly suggest you don't try to implement this yourself.

@lukechip:

That fsutil query just tells you whether the file system is sending down trim commands or not. The file system doesn't know (or much care) what kind of storage lies at the very bottom; it might even be multiple types (think volumes that span multiple disks, RAID arrays, etc.). If trim is enabled, NTFS sends down trim commands on all volumes and lets the underlying layers sort it out. I'm not sure how you can get the physical characteristics you want about your SSD drive. As a start, poke around at its properties in Device Manager (devmgmt.msc).

re: TRIM and OS cache/buffer 'flushing'

me&er — Wed, 13 May 2009 12:45:41 GMT

<<QUOTE>> @me&er:

Tuesday, May 12, 2009 3:37 PM by craigbarkhouse<<QUOTE>>

Many thanks Craig. Appreciate your responses.

Just to put the record straight, I’m not looking at programming an alternative to TRIM and have no wish to compromise

I’m using a propriety TRIM tool in beta at the ‘mo from OCZ/Indilinx for their Vertex series with the Barefoot controller. If you guys haven’t got it yet.. I would give it a go. I’m not marketing it at all.. all I am interested in is confirming that it actually consolidates the free space effectively and without excessive ‘overhead/nand longevity issues’. My limited testing ability confirms it initiates fine in cmd/conhost.exe and works along the FS stack with ‘native’ Microsoft storage drivers, writing to the MFT and paging to file effectively with no file corruption identified.. the end result being a noticeable increase in random read/write speed at little or no cost to sequential read/write. Calculations for the longevity of the specific MLC drive is somewhat complex but nevertheless quite acute on the 30GB models, so I will need to compare this propriety tool ‘initiated’ usage against the FS one.

TRIM or trimming or defragmentation/consolidation of free space is much the same for my needs.. what is important is fitting this in to how it effects the smaller capacity MLC SSD, as part of a write optimisation strategy.

This strategy involves looking at controller/FW Wear Leveling and how it interacts with Win 7 with both SLC and MLC SSD which is why any form of TRIM and Device/OS write cacheing is right at the top of my analysis. Neatly brings me onto this question:

If..

<<QUOTE>>for devices that do cache writes in volatile memory, Windows 7 expects flush commands and write-ordering to be preserved to at least the same degree as traditional rotating disks...<<QUOTE>>

Q. What’s the reasoning behind Win 7 default ‘enable’ of OS cache/buffer 'flushing' as opposed to the default ‘disabled’ in Vista/XP (albeit described as Advanced Performance afaik).

Regards,

re: Support and Q&A for Solid-State Drives

tremelai — Thu, 21 May 2009 06:53:32 GMT

I have noted that W7 is massivly faster on my SSD's than XP. In my research, I found that Windows XP/2003 have three major drivers. (scsiport.sys, atapi.sys and storport.sys)

I have found that SSD's fly under storport RAID controllers and don't do as well under atapi.sys.

I understand that the old scsiport.sys and atapi.sys use a form of the C-LOOK disk scheduling algorithm while storport does no software scheduling. (leaving scheduling up to the controller.)

Q. Since SSD's do not have moving parts, is the lack in C-LOOK overhead accounting for the performance increase with storport?

If so, what scheduling changes, if any, does the new ataport in windows 7 undertake when an SSD is detected?

Regards

ocz core v2

venomz3 — Thu, 28 May 2009 22:47:57 GMT

My ocz core v2 ssd isn`t detected as ssd. scheduled defragmentation is enabled to that drive and also superfetch is enabled. So i presume the windows hasn`t detected drive correctly. Hdd is attached to jmicron jmb363 controller and is confugured as ide because ocz page said that drive should no be configured to achi mode. and only settings in jmicron are raid,ide, and ahci.no sata mode without ahci enabled

SSD not detected?

mdsharpe — Fri, 29 May 2009 12:00:40 GMT

Fascinating blog post, thanks very much! I have a question though regarding the SSD in my Acer Aspire One A110.

It does not appear that Windows 7 (build 7100) has detected the SSD, as defrag and superfetch are all still enabled.

Is there any way to force Windows to see a drive as an SSD?

Thanks. Matt Sharpe

re: Support and Q&A for Solid-State Drives

livelock1000 — Sun, 31 May 2009 18:13:59 GMT

Fascinating indeed, though I have the same problems and questions.

I have installed Win7 RC on a Samsung PB 22 J, which is en par with Intel X25-M in most benchmarks (except #IO) and exceeds in write performance.

Cryistal DiskBench gives me for example

Seq Read/Write 210-220MB/s / 80-160 MB/s (Avg 140)

512k 170MB/s / 100MB/s

4k 15MB/s / 6MB/s

...the 4k category is the only where x25-m seems to have better results up to 25 / 50 MB/s.

Now here is what Win7 did:

- Performance Indicator stuck at 5,9 (according to your explanation there must be one deficiency according to your tests)

- Superfetch: enabled

- EnableReadyboost: 3 = enabled

- EnablePrefetcher: 3 = enabled

- Defragging: Scheduled and possible

- Alignment: no clue

Clearly this is one of the fastest drive so I conclude Win 7 did not notice this somehow.

1. How can what Win 7 did recognize, how do I see the tests Win 7 did and used to decide this is not a SSD?

2. How can I force Win 7 to detect this as a SSD?

3. How can I tell Win 7 while installing that this is a superfast SSD to enabled SSD alignment?

Regards

re: Support and Q&A for Solid-State Drives

livelock1000 — Mon, 01 Jun 2009 20:08:50 GMT

I have bonus question:

EWF (from XPe) & MS Steadystate seem to sequentialize random writes, I read somewhere that such a device is in Win 7 already but not yet functional? Is this true, what is it? what´s the name? can it be activated?

This sounds like a must for SSD drives.

re: Support and Q&A for Solid-State Drives

hedges — Sun, 07 Jun 2009 21:59:30 GMT

This new SSDs read , expected more than than 200 MB/s? Thats quite good enough

Trim for Raid 0

PeterSt — Wed, 17 Jun 2009 01:26:09 GMT

i have 6 SSD's in Raid 0, working fine, but have noticed degradation in performance. As of yet havn't seen definitive statement from MS re Win 7 modding Trim so SSD's can be optimised whilst in Raid 0 array. (using Asus P6T deluxe mobo, Intel i7 cpu) Have heard lots of encouraging talk, but no firm word yet. Anyone heard any?

re: Support and Q&A for Solid-State Drives

Homeschool Curriculum — Wed, 17 Jun 2009 12:40:01 GMT

re: Support and Q&A for Solid-State Drives

Homeschool — Wed, 17 Jun 2009 12:42:25 GMT

So i presume the windows hasn`t detected drive correctly. Hdd is attached to jmicron jmb363 controller and is confugured as ide because ocz page said that drive should no be configured to achi mode. and only settings in jmicron are raid,ide, and ahci.no sata mode without ahci enabled

re: Support and Q&A for Solid-State Drives

Online High Schools — Wed, 17 Jun 2009 12:45:42 GMT

By having the file system send down trim commands when appropriate, you enable the drive to immediately benefit from this information. There's very little overhead to these commands. Contrast this with defragging, where if you were constantly defragging everything the cost would outweight the benefits. I strongly suggest you don't try to implement this yourself.

re: Support and Q&A for Solid-State Drives

must university scam — Wed, 17 Jun 2009 12:59:13 GMT

the ability to support trim, since our other port drivers don't implement trim. This may change as the market evolves. I don't know if any 3rd-party storage drivers implement trim as of yet, but yes, they would have to implement it for it to work.

re: Support and Q&A for Solid-State Drives

PeterSt — Sun, 21 Jun 2009 04:01:40 GMT

this can get a bit mumbo jumboy for me, but are you saying that that if the raid controller can address this "rotational speed as zero for the units of storage (disks)" then windows will see the raid 1 array as i disk and the Trim tool in win 7 will work on a raid array?

re: Support and Q&A for Solid-State Drives

venomz3 — Mon, 22 Jun 2009 19:24:22 GMT

So I presume that standard pci ide controller doesn`t report rotational speed to windows because ocz core v2 wasn`t detected as ssd.

Maybe I shoud give windows 7 jmicron drivers when installing not afterwards. Microsoft shoud be more informative with these post.Example information about hdd detection

and information about reading and writing changes in low lever.I would very interesting and I guess many people who read this blog.

Has it have to be manufacturer drivers or manufacturer supplyed to microsoft drivers or is it ok to install with standard drivers and expect windows to detect drive as ssd ? I usally give windows drivers afterwars if windows detects drive so I can install to it.

re: Support and Q&A for Solid-State Drives

Silver0066 — Tue, 30 Jun 2009 20:10:52 GMT

Using Windows 7 RC, I cloned an SSD drive back to my Raptors and removed the SSD from the system to put on another system.

Since this system no longer has SSD drives, is it advantageous to reset DisableDeleteNotify to 1?

Many thanks for all of the information.

re: Support and Q&A for Solid-State Drives

robinanil — Mon, 06 Jul 2009 19:17:33 GMT

Hi Steven, I was wondering what kind of compression codec does NTFS use. For example on the internet i have heard of people saying the overall transfer rate decreased while using compressed files. I am sure you would have heard of the LZO Algorithm http://en.wikipedia.org/wiki/Lempel-Ziv-Oberhumer

Have you guys implemented this on NTFS. The speed and performance gains on a pentium 133 as given here http://www.oberhumer.com/opensource/lzo/#speed

is mind boggling.

I have been waiting for such a solution. Awaiting your comments

re: Support and Q&A for Solid-State Drives

webtasarım — Tue, 07 Jul 2009 15:46:07 GMT

Thank you for preparing such an article is good and my long

re: Support and Q&A for Solid-State Drives

phantomd — Tue, 28 Jul 2009 12:51:29 GMT

I start to test my solid state drive on Windows 7 64bit Build 7201. With Windows XP and Windows Vista i made very bad experience, after a couple of weeks, the solid state lost a lot of speed and simple internet browsing was terrible. I checked after installation in Windows 7 that Disk Defragmention is off and also Super Fetch is disables. Disk Defragmention was on Manual after Installation, i have Disbaled now. Super Fetch was active and start at the boot. I have disabled now. I use a 32GB OCZ Solid State. Is there already a tool aviable to check which ssd drives are certified ?

web tasarım

kymk — Wed, 29 Jul 2009 21:42:06 GMT

Thank you for preparing such an article is good and my long...

http://www.parcakontorbayiniz.com

re: Support and Q&A for Solid-State Drives

condos — Mon, 10 Aug 2009 08:54:47 GMT

Little bit confused about NTFS code compere but no doubt you written best article

re: Support and Q&A for Solid-State Drives

condos — Mon, 10 Aug 2009 08:56:25 GMT

yes but sure wanted to know more about if you specify about NTFS it would great help by yours.

re: Support and Q&A for Solid-State Drives

diploma-2009 — Thu, 20 Aug 2009 20:52:45 GMT

Thanks for the article. Even though I'm not the greatest fan of Microsoft products, I'm pleased to know that Windows 7 supports SDD. As far as I know, the technology is quite new, quite expensive and I don't know any real people who are using SDD. However, I think that it is really promising and I'm waiting for the moment when I can try it myself.

re: Support and Q&A for Solid-State Drives

ilyas_yuregir — Thu, 05 Nov 2009 11:08:27 GMT

Ankara hosting, ucuza hosting, ilyas, ilyas yüregir, ankara web tasarımı

re: Support and Q&A for Solid-State Drives

cash gifting — Fri, 13 Nov 2009 23:33:59 GMT

yeah its quote helpful with the ssd support im very supportive of that

re: Support and Q&A for Solid-State Drives

web-tasarım — Thu, 19 Nov 2009 22:59:48 GMT

I congratulate the family of 7 products microsoft windows vista windows than to high-performing and visually richer ... Congratulations windows7.

re: Support and Q&A for Solid-State Drives

web.tasarım — Tue, 01 Dec 2009 10:39:27 GMT

Thanks for the very informative and enlightening documents. Thanks to myself every day for more of you.