Talk:Hybrid drive/Archive 1

Untitled discussions

This is an emerging technology of NAND flash and HDD, the hard drive companies engineers should be looking at the discussion pages like these to answer consumer's doubts. Also can everyone spell check and grammar check, it looks childish to misspell technical words, or something simple such as disadvantages or deteroriates(deteriorates). Also much of the article is not updated to March 2007 recent news articles such as the Samsung 160 GB/256MB NAND buffer drive for $40068.161.125.123 16:36, 17 March 2007 (UTC)Lyskar

This article does not address any of the possible flaws concerning the design of this technology. As it is an emerging thing, I don't expect all the details to be hammered out, but I see flaws on fundamental levels.

Modern operating systems and hardware organize memory hierarchically. This is usually done such that the smaller, faster tier of memory above a memory level contains a subset of the current tier. The reverse is also true; the contents of any level of the hierarchy is replicated in the level below it. If the cache below any level is too small to contain all of the data at a higher level, all accesses to it (that were not cached at the larger, higher level) will be misses if both schemes use temporal locality. (This can be mitigated somewhat by prefetching.) Thus, since operating systems support disk IO buffering using RAM and since most systems support large amounts of RAM, most if not all cache misses on the RAM will also miss the FLASH memory and require a magnetic disk access. Thus it does not provide as significant an advantage as it initially sounds; for the first four benefits, the disk should already be spun down.

Furthermore, FLASH technology is slower than magnetic disk technology in terms of write throughput speed. According to Samsung's website OneFLASH (the technology cited in the external article) is capable of ~10MB/s of sustained writing, whereas a typical 7200 rpm hard drive is capable of at least 40MB/s sustained. While reading is faster at 108MB/s, it has to exist in the FLASH before it can gain this benefit, which goes back to the previous problem; a disk access penalty is taken.

You're forgetting that the FLASH inside the drive doesn't have to be in a single bank but in a rack of sorts where eg. 5 FLASH drives read and write concurrently effectively multiplying the read and write speed. All it needs is multiple flash banks and separate caches for them: Expensive, probably, reliable, probably not, possible, yes. - G3, 07:56, 10 November 2006 (UTC)

In addition, while a magnetic disk write is capable of 40MB/s sustained, it is much slower for random access reads. This is where FLASH is much faster, as it's random access seek times are essentially zero. The question is: are sequential or random reads more common? Depends on the task. Video editing -> sequential, most other things -> random.

No, this previous paragraph completely misses the point. If it's random access, it's either already in the RAM (which is *far*, *far* faster than any current Flash drive) and doesn't need to be read from the hard disk, and if it's not, then it's probably not on the Flash storage either. For sequential reads, caching in RAM will take care of the vast majority of reads, and the Flash part of the hard drive does not really aid much. The only case where it is indeed very useful is for machines with very limited amounts of RAM.

I agree. Think of it as a drive with a huge cache, write back enabled, and set to spin down very quickly. The main reason behind it is, I think, not so much the unique features of Flash, but the fact that disk drives still grow very quickly in capacity, and commensurable buffer RAM is expensive and needs a lot of power. I don't see the issue of write data rates - the Flash buffer would only have to sink the data until the disk has spun up, then it could act as an ordinary write buffer (when the RAM write back buffer overflows.) So, bottom line, it can't get any slower than current drives (as soon as the platter has spun up) because the data doesn't *have* to go into the Flash. As for read caching, a GB is probably more than your usual Windows based laptop uses for caching, so it should give a better hit rate. My machine has 2 GB and Windows uses "only" 600-odd MB for caching right now. Ralf-Peter 01:18, 6 December 2006 (UTC)

Regarding instant boot, if the flash is used as a cache, which the wiki article suggests, the OS would need to be preloaded to the flash at every shutdown, unless it had not yet been cleared. It simply delays the time, not remove it.

You could always keep the first X blocks in Flash. But yeah, all those system DLLs will probably get displaced sooner or later. Unless you have special commands... Send "LOCK EVERYTHING FROM NOW ON IN FLASH" from the BIOS, then turn it off once the boot phase is over. Ralf-Peter 01:18, 6 December 2006 (UTC)

• Patterson & Hennessy Computer Organization And Design. 3rd ed. Elseveir, 2005.
• OneNAND product information

tw 20:08, 15 June 2006 (UTC)

That's valid material, and sounds like you've done your research thoroughly. I suppose we could add in another section detailing those concerns, or possibly a pro/con comparison with regular disk drives. But I suppose we really won't know too much about things until the drives actually come out in a year or so.

Still on the subject of disadvantages: What would happen to the hardrive when the flash memory "worn off"? --Pinnecco 07:53, 4 August 2006 (UTC)

Whether intentionally or not, parts of this article read like a promotion, so I changed the language a little. Also, put things in present tense, as these drives have been developed, even though they haven't been marketed (I understand my reasoning may be a little shaky, feel free to disagree). Last, the "Advantages" section is now an ordered list. Sloverlord 18:45, 23 September 2006 (UTC)

I don't see how this article is controversial / not neutral. It's new technology, so perhaps some of the benefits will not be realized, but this can be fixed by changing statements about the present to the future. (For example, change "This offers benefits" to "This promises benefits", and adding a disclaimer at the top of the "Benefits" section that "Because this technology has yet to ship, it remains to be seen which of these benefits materialize." or something like that.) The primary benefits of flash for this application are fast random writes (for enterprise workloads), less power/heat/noise (for laptops and similar), and fast booting for consumers (hard drives limit boot time to spinup time -- see LinuxBIOS). It's also cheaper than RAM -- I could imagine copying movies to flash and playing with the drive spun-down. Given that random reads require knowing what will be read in advance (unlike writes), I don't see a big benefit. Also, it's premature to claim better reliability. - James, 26 November

here is some nice data about the effect of adding a couple hundred MB of flash to a system - and it looks like it works (although they are using a laptop with very little memory)

DeFrag

The Arcticle does not talk about disk defragmentation. I could see that with this technology The hard drive could automatically defragment as you work. Is there any work in this area? Zginder 22:33, 3 November 2006 (UTC)

Kind of, many file systems used by Linux don't fragment files. —Preceding unsigned comment added by 76.191.180.179 (talk) 05:36, 8 April 2008 (UTC)

Defragmentation is a file system thing. Nothing the disk drive can do by itself. Unless you want to implement NTFS on the drive. Ralf-Peter 00:55, 6 December 2006 (UTC)

Theoretically, it could speed FS defragmentation. An extra 1GB of memory close to the drive gives you a lot of room to play with as you shuffle things around. A bigger cache can also mean less fragmentation, if the kernel is aware of it. But I don't think these possibilities are notable, not until they've been proven. --Nasarius 21:18, 20 December 2006 (UTC)

The hardware itself couldn't do the defrag on its own. But I think that what Zginder means is that a defrag application could be optimised to take advantage of the flash chip. --Pinnecco 18:00, 15 November 2007 (UTC)

Flash Write/Rewrite limit?

Isnt there a limit to the number of times you can write and rewrite data to a flash drive? Pdinc 13:43, 10 November 2006 (UTC)

Yes there is, read this: http://ask-leo.com/can_a_usb_thumbdrive_wear_out.html There must exist some solution to this otherwise the hybrid drives will only last a few years (and gradually turn slower). --Sire404 08:38, 22 November 2006 (UTC)

Not completely sure, but I got this from the Flash Drive article - "Modern NAND-based flash drives often last for 500,000 or more erase/write cycles." Paddyman1989 12:50, 16 November 2006 (UTC)

Even if true, 500k is not as much as it seems for cache-like performance profiles, which is the touted benefit of the drives in the first place. I would be wary of buying such a drive until 3rd-party tests under typical (and non-typical) usage reveal their true lifespans. Eriol Ancalagon 05:44, 2 December 2006 (UTC)

Keep in mind, that the flash is usually hidden behind a "virtual memory" controller that automagically rotates write usage. That means the larger your flash (cache/usb stick/camera card) is, the longer any given cell can hold out. If you use it just to store annoying filesystem seeking pains: directory structure, file locations, integrity journals, it could last a long time. How often do you turn your directory tree upside-down? Such a usage would need a filesystem update (I think some linux filesystems are halfway there with seperate journal support). 68.0.226.163 02:24, 18 July 2007 (UTC)

Commercial article?

This article reads more like an ad than an encyclopedic article. Where are the possible disadvantages of this type of drive (including data corruption on power loss, rather limited write/erase cycles, higher cost)? I would also like to see few words written about compatibility with different OSes and older systems. I NPOVed this because of the lack of critique or, indeed, neutrality. - G3, 07:45, 10 November 2006 (UTC)

Flash Memory

We could simply provide a link to the flash memory wiki so people can see all the limitations of the technology.

Disadvantages

While this article discusses many of the "advantages" of the usage of flash media as an intermediary, no attention has been paid to the service life of the media itself. I wouldn't want a new harddrive that i would have to perform an overhaul on every year. The longevity of flash memory is only about 1,000,000 writes per block on the best media, and it's far less on others. Where hard drives are not intended to be thrown out like a common thumb drive every few years, the use of flash media, even with a wear levelling (-leveling) component, may significantly reduce the useful life of these hardrives. I think the advantages section needs a counterpoint of disdvantages(disadvantages)...

This still reads as though it is a promotion of the technology rather than a factual accounting of it.

What about the number of start/stop cycles? Yeah I know these days we have ramp load in most notebook drives with several 100,000 cycles, but still. Ralf-Peter 00:58, 6 December 2006 (UTC)

Microsoft failed so far to separate read-only components such as bootloader, kernel (and possibly kernel modules) from the regular filesystem AFAICT. Unices can boot from flash, and appliances with other OSes (think of mp3 players, routers, etc.) also often boot from flash. The advantage they use is that such mostly read-only data is not something that is updated every day, hence the flash media is not worn out too much. Someone having an idea for a hybrid drive is just driving another marketing hype. I mean, what would be the implementation like? (1) Does a Hybrid drive show me two block devices, e.g. /dev/sda and /dev/sdb -- one for the flash area, one for the platter area? Then this "Hybrid drive" would just be an enclosure around two today's drives (CF+disk), to put it bluntly. (2) Does a Hybrid drive appear as one, /dev/sda? If so, how do you control what goes into the flash area and what does not? By use of more ATA commands cluttering the standard? No thanks, then I'd rather go with (1). Oh and BTW, if booting is really _that_ slow that we need flash-driven hard disks, it is time to change OSes or remove some crap. j.engelh 00:45, 10 December 2006 (UTC)

Is this right?

The second instance is when the user must access a new file from the hard drive that is not already stored in the buffer. In this case, the platters must spin up to access the file and place it onto the buffer, whereupon the platters will once again return to an off state.

I don't know anything about this technology, but logically this seems wrong. When a file is read off disk, normally it ends up in the hard disk's (volatile) cache, as well as in the computer's main (volatile) memory cache. There's no particular advantage in filling up the (non-volatile) "buffer" with read only data. Can someone verify this? Stevage 02:18, 22 November 2006 (UTC)

I think the advantage is that the buffer is much larger than traditional disk caches, and that it can be used for writes as well as reads, since it's non-volatile. David McCabe 00:36, 25 November 2006 (UTC)

For a computer that is up 24/7, there is almost no advantage to a flash-based cache over ram at similar sizes, save cost. At similar price points, a flash-based cache should allow for larger cache sizes compared to increasing system memory which can significantly increase performance if the larger cache allows for eliminating any disk access for the item cached. The major advantage seems to be for laptop users who do complete shutdowns frequently rather than using a hibernate feature. For those usage patterns, taking advantage of temporal locality across multiple computer boots can be a significant advantage.

Biased toward Windows

computers using hybrid drives may be able to achieve extremely fast (under 10 seconds) or even near-instant boot up times.

Microsoft Windows is one of the few beasts that need more than 10 seconds to boot-up. Under 10 seconds is in no way _extremely_ fast, given the technology and computing power available today. Dont let computer users think Windows' slowness is the standard.

If you make a modification to make your your VW beetle do 0-100 in 5 seconds, I would call that "extremely fast". Doesn't matter what a ferrari or McLaren F1 would do...Stevage 23:22, 22 November 2006 (UTC)

Arguing semantics I know, but, I would reason also that it is not `extremely fast,' by your analogy the McLaren F1 is extremely fast (I'm sure 0-100 in 5 is no easy feat - but what about corner speeds), while the VW has received a significant performance boost, or any other language that describes a relative rather than absolute speed differential. -- —Preceding unsigned comment added by 24.221.12.89 (talk • contribs) November 26, 2006 @ 17:14 (UTC)

Hm, my Fedora 6 linux laptop takes 1 min 29 seconds to load (from the moment when I select linux in GRUB to the login screen). -- Convex hull 02:48, 27 November 2006 (UTC)

Perhaps another distro is best? Archlinux with all sorts of services boots in about 30 secs on an old 2.4 GHz laptop.

when writing the initial statement, i was thinking about my QNX 6 (www.qnx.com), which takes between 10 and 15 seconds to boot up on a 400Mhz Celeron, from the moment i press the switch to the login screen, half this time for the BIOS, the other half for the OS. the OS is complete with GUI, network, browser, etc. everything a basic user would need to use a computer. Another example, a very lightweight DOS install on a 8MHz 80286 takes 5 seconds to boot, but this time, you dont have anything fancy (you dont have anything at all in fact)...

OK, I'm sure there are lightweight OSes which boot in under 10 secs from a regular drive. However, stating that "Microsoft Windows is one of the few beasts that need more than 10 seconds to boot up" is just silly. -- Convex hull 07:47, 2 December 2006 (UTC)

QNX is far from being 'lightweight'... just more efficient, and if they can be that efficient, every other OSes should also. As a side note, i just realized: QNX and every other OSes i saw booting in less than 10 seconds used microkernels, Windows and Linux and other OSes i saw booting in more than 10 seconds are MACROkernels. anyway, we are getting far off-topic. —The preceding unsigned comment was added by 81.240.46.231 (talk • contribs) 10:18, 3 December 2006 (UTC).

There seems to be another problem, the license: The Samsung Hybrid drives are shipped with this note:

'Hybrid Disk Drive products are licensed for use only on devices that deploy the Windows VISTA operating system as the principle operating system. If you or any other party install(s) an operating system on the computing device that is not Window Vista, the use of this Hybrid Disk Drive may require an additional license from Microsoft. For further information, please contact Microsoft.'

Vista is at this moment the only OS that supports hybrid disks, but this note seems to state that you may not use these disk with another OS?? Allso see: This link —Preceding unsigned comment added by 80.126.30.223 (talk) 11:24, 30 September 2007 (UTC)

Improved Reliability?

I think the whole Improved Reliability thing is a complete bunk, at least on desktop computers. The most deteriorating thing a hard drive normally has to do is to spin up the platters which is effectively equal to a cold boot. For example (data taken from Samsung's random laptop HDD page):

• - Linear Shock (1/2 sine pulse): Operating, 2ms 325 G, Non-operating, 1ms 1000 G

2ms x 325 Gs worth of acceleration is roughly equal to dropping the HDD from a height of 2 meters and that's without any case absorbtion/deformation, while 1ms x 1000G means 4.9 meters. For some reason, Samsung doesn't report start/stop cycles for laptop HDDs but browsing through various desktop HD Ds shows the following data:

• -Start/Stop Cycles (Ambient) 50,000

50,000 start/stop cycles will normally happen in 50,000 boots = a lot, but if the HDD has to spin up, say every 20 minutes, then it's only 16,000 hours of operating time which roughly equals around 2 years - in optimal starting conditions. Not to mention the fact that while spinning up the HDD uses much more power than in normal running mode resulting in extra stress to other components (including laptop battery) and will halt all other actions in Windows XP unless you have dual-thread (core/ht/2 processors) capable computer. - G3, 04:30, 13 December 2006 (UTC)

Infact, the only place where having a hybrid drive improves reliability is on a mobile platform with sporadic needs of HDD storage (eg. car computer but not one which needs constant HDD access like data logging). In all other cases hybrid drive deteroriates the reliability (well, except if you move your desktop computer around a lot or use a laptop while eg. bungee jumping). - G3, 04:52, 13 December 2006 (UTC)

Power Failure

What happens if the system, laptop or desktop, suffers an acute power failure? Normal HDD has enough kinetic energy to retract read heads to safe zones and only the stuff that was being written was lost...However, what happens with a hybrid drive? Which data takes precedence the data on flash or data on actual HDD if the power is removed without cache-flush? What if power is cut during cache flush? How badly can acute powerloss corrupt the drive (eg. MP3 players unplugged during writing oft require a full reformat)? Is there a backup battery or additional cache level to sort these things out?

I would like to see this article have facts and answers instead of hopeful fiction - G3, 06:05, 9 January 2007 (UTC)

We're just guessing

Guessing is probably the wrong word. But what I've read on this page sounds to me like a lot of intelligent people discussing a new implementation of existing technologies, none of whom have any direct information about this new implementation. Other than Dvorak, who insisted that Vista won't run properly without one of these, I haven't read anything on them other than press releases and corporate promises. Are there white papers or better than that somewhere that I'm not seeing? Kisc 16:41, 3 January 2007 (UTC)

Yes, this is all guess work and it is something that should be noted in the article in very bold letters. Improved reliability, performance, etc.. values are meaningless commercial adwords without independent testing to back them up. If this technology is buffed by Microsoft then, I fear, we'll have to wait for a long time for such a test to emerge from an independent source. But, as Wikipedia tries to be encyclopedia we should add the necessary weasel words to the article to make sure the reader understands that the sources are either unreliable, unverifiable, biased or, as is the case, non-existent. - G3, 15:23, 4 January 2007 (UTC)

Faster Performance?

Can someone explain to me how does Flash memory cache on HDD improve performance over cache on RAM?

Secondly, can someone - again - explain to me how Flash memory *cache* on HDD improves boot times? For the operating system to boot from Flash it has to reload itself on to the cache every time the computer is being shut down - What this equates is to slower shutdown time when critical components are copied from the HDD to the flash storage OR alternatively there's a reserved space, a flash drive if you will, within the Flash memory for the critical operating system data, in which case the "increased boot performance" is because of the logical SSD and not because of the Hybrid aspect ie. you're better off with a 2 gig pendrive (or similar) and a fast HDD than a hybrid drive.

There’ s no performance improvement with caching on flash memory, over SDRAM of any kind. This improvement is probably achievable in case of a computer in sleep or hibernation mode, where the platters have spun down. In a usual, contemporary computer, featuring a non-hybrid hard drive, resuming to G0 from S3 or S4 state premises that the hard drive will be spin up its platters, to deliver the appropriate data, and, finally be available to its controller. This procedure takes some 3-5 seconds for the hdd to spin up the platters and initialize its BIOS. From the other hand, a hybrid hard drive should be expected to react in a different fashion, as it can store these data in the NV cache, hence, no platter spin up would be necessary, until the system needs data which are missing in the NV cache, and, so, they must be read from the platters. This mechanism does not make a big difference, but, off course, there are some changes to be performed in the h-hhd BIOS. In addition, I would say that longer shut-down time is not an issue, as it happens in a not-in-use time window.

Thirdly, even with faster read/seek capability SSD drive will not boot (Windows XP) much faster because the boot sequence includes much more than reading data from HDD to memory. See eg. this for a real world comparison (5 seconds off of 30 seconds compared to - a slower than desktop drive - laptop HDD). - G3, 05:52, 9 January 2007 (UTC)

Recovering from Flash failure

Is there any reason (other than marketing) that the flash in the drive could not be removable/replaceable? Think of that blank spot on the front of a computer case that the hard drive sits behind. Why can't this be a door where the user could open and remove the flash portion of their hard drive and replace it? This allows for replacing worn out flash memory but also for the user to optimize the flash portion of their drive for games, rendering, database, web serving, cash register, or whatever task they do. The amount and type of memory could be matched to the current task. --Zerothis 09:05, 20 January 2007 (UTC)

I was thinking exactly the same. But that's a question that the manufactures should answer, really. There would also be the massive advantage that users could in theory choose the flash cache size for the disks. But then I am sure that companies would manufacture yet another flash format for such disks -- as if SDs, xDs, Memory Sticks, etc is not enough -- and inflate the prices! --Pinnecco 17:41, 15 November 2007 (UTC)

RAM Drive?

Isn't this just the old idea of a RAM Drive but using dedicated memory? --Zerothis 09:05, 20 January 2007 (UTC)

Verifiability and Reliable Source issues

This article has extreme problems with verifiability and reliable sources. Hardly any of the content is verifiable, and all of the sources listed appear to be paraphrases of content from manufacturers. This is in contradiction to Wikipedia policy and guidelines. While the subject is certainly notable, we need to think about how to cover it properly in the context of a complete lack of the reliable sources expected for a technology. One idea might be to cover it from a more media-oriented perspective, since little is actually known about the technology and most of the notability is due to coverage in the popular media. I might work on a rewrite of the article from this perspective in the next few days. --Constantine Evans 11:52, 25 January 2007 (UTC) Hardly any of the

Verification

Verification of the statements in the articles requires access to technical reference manuals for the discussed hard drives. "whereupon the platters will once again return to an off state" can only be verified from the technical reference manuals. "the head should be docked most of the time" is an inference, given that most, if not all, modern hard drive manufacturers state their drives automatically park (dock) when the platters spin down (sleep or off mode). "A hybrid drive requires spin-up and spin-down more often than a normal hard drive, which is often spinning without a break" - this is partly implicit, if you verify the platters spin down for cache-only mode, then it requires spin up and spin down when data use exceeds 1GB (or the cache size), while I've never seen a hard drive that sleeps or spins down as part of the natural duty cycle, without power management intervention (The hdparam utility - http://prdownloads.sourceforge.net/hdparm/) has historically allowed the user to set sleep timeouts for hard drives.) As any hybrid hard drive is going to spin down until 1GB of cached data is used up, it would be implied, but not supported by research data, that it would spin up and down more often.

As far as wear-and-tear being worse for a constantly cycling (spinning up and down) drive, I've never actually seen a research articles, just references to them. But starting friction is the common war cry there. Your mileage may vary.

Here's another big point, I didn't see mentioned - windows, by default, implements an aggressive memory caching algorithm, for the swap space (commit charge.) Now, they'd have to disable caching for the swap file, or not swap so much. My commit charge is currently 695MB, and if I were to run a search, spybot s&d, an av program, or anything like that, my commit charge goes way up over 1GB. So, just starting my computer and having some automated virus scan, running a large file-text search, runs up the commit charge.

Windows writes dirty pages constantly, that alone should use the write cycles on flash ram within a couple of years, not even counting what the user does with their data files.

I think the article is poorly written, and should be deleted. I was surprised to see it here, and thought "advertisment" when I saw the link on google (I'm looking for user-reported failure of samsung drives, and other drives, trying to research the most stable drives at this point in time.) Earlier, I was looking for articles on phytic acid in soybeans, and the only thing available was a slightly biased wikipedia article, copied across 3 google results pages! It's really besides the point, but google gives too much weight to wikipedia.

I found the talk page far more informative food for thought about these drives than the actual article. Swap the two? - just joking. I hope my blurb is helpful for whoever rewrites the article, I'm not up to it - too technically oriented to disk manufacturing. Rhodescus 20:24, 4 February 2007 (UTC)

Availability

I am wondering when these drives are going to be available? Also if there are any plans from the major computer suppliers, i.e DELL, HP in including these drives?

Thermal Cycling and other issues

Spinning the drive down and up poses serious reliability issues. One is that it will result in significant (20 celsius degrees) variations in drive temperatures in a typical PC. These thermal variations are likely to cause more failures than operating the drive at a constant temperature. Spin up stresses the motors, the motor drivers, and the power supplies, as has already been mentioned. Spin up also carries a risk of the head dragging on the platters unless the drive design physically holds the heads up when parked (though there can be issues of the heads falling when released).

Swapspace can potentially lead to significant wear on the flash memory. To work around this, the drive would need to support additional commands and the OS have drivers that implement those, to identify ranges of swap used to the drive (i.e. swap partitions and swap files which may be fragmented) so the drive can cache those in RAM, not flash. Swap does not normally need to be preserved after a power failure; it is normally discarded at boot and while there might be some data there you might like to recover it is likely to be incomplete and disorganized so that in practice it is very rarely used. However, swap usage then interferes with spinning down the drive. Also, reads should really be cached only in RAM (either on drive or in system - to minimize flash wearout) and writes only in Flash (crucial for recovery). A hard drive is usually read far more often than written. If the flash is well implemented and the RAM isn't reduced, wearing out of the flash should mean that the drive degrades into a standard drive rather than simply failing. However, power dissipation will go up and/or there will be more spin up/spin down cycles.

Since OS support is likely needed to minimize flash wear through swap usage, there is actually little advantage to putting the flash on the drive itself as the OS can be configured to use flash anywhere on the system. It appears Microsoft readydrive already supports flash separate from the drive as I have seen USB thumb drives advertising readydrive compatibility. On drive flash may have a faster parallel interface, however. OS modifications to more tightly integrate the flash with the filesystem journalling system are also probably called for.

It is also worthwhile to compare the benefits to increasing system RAM. 1GB of extra ram and a small amount of flash (to buffer writes) would probably give much better performance (though a small flash will probably increase the number of spin up/spin down cycles). If you add more system RAM, you can eliminate swap usage and cache reads more. It is a question of whether the drive would wear out faster due to spin up/spin down and thermal cycling or due to flash wearout. Adding cheap, slow, low power consumption RAM and flash to the motherboard along with OS support would probably have the greatest benefits in the long term.

Someone mentioned the flash memory being less recoverable than data stored on the platters in the event of failure. Presumably, this is because data recovery operations often involve swapping out the electronics. However, if the flash memory has a JTAG interface, the contents will often be recoverable even when the electronics board is dead.

An interesting option would be to run the drive at a small fraction of its normal speed instead of powering down. However, there are aerodynamic and signal amplitude issues.

While the flash may have advantages in terms of battery life, the system will still have to be designed to handle the power and heat dissipation requirements of the drive operating at full capacity.

There is a good chance that microsoft will (or already has) lock part of the OS permanently in the flash (reducing data cache) in order to disguise slow OS start up time, just as they now pretend the system has booted before it actually has and pre-load office into memory to hide its true startup time at the expense of other applications. Whitis 02:43, 22 February 2007 (UTC)

Inaccurate

"..with Samsung introducing their first drives in January..."

Its now march and no drives from Samsung. —The preceding unsigned comment was added by Gibsondj (talk • contribs) 21:39, 1 March 2007 (UTC).

Move to Hybrid disk

This article should be moved to Hybrid disk. From the Hard disk article: "Strictly speaking, "drive" refers to a device that drives (removable) media, such as a tape drive or (floppy) disk drive, while a hard disk contains fixed (non-removable) media." - Face 11:26, 2 March 2007 (UTC)

Sentence Ambiguity

"Lower reliable lifetime - Flash based memory has relatively limited read/write cycles compared to hard drives which can last over a decade without mechanical malfunction."

I assume that hard drives "can last over a decade without mechanical malfunction". But since this is an encyclopedia article, I have to say it is not very clear whether it is Flash based memory or hard drives that "can last over a decade without mechanical function".

Flash has a similar MTF when compared with hard drives

Folks, many studies have been done. It's not 1999. Hard drives have a finite number of writes per sector as well, that's why there is a huge "backup" of redundant disk space. Every drive has 20% more space than you can actually see to accomodate this. Flash is the same way. Flash memory is just as reliable as a hard drive within the warranty period. Outside the warranty period, perhaps there is additional risk with flash - but seriously, anyone who uses a hard drive that long is crazy! —The preceding unsigned comment was added by 64.115.129.226 (talk) 21:31, 20 March 2007 (UTC).

I have a 20 meg HDD from early 90s, 4 GB drive from mid 90s, 80 gig drive from late 90s (in active use, Maxtor even, unreliable brand to boot)...All in perfect working condition last I tested: The only malfunctions I've had were because of mechanical drive failures, not because of any damage to the magnetic media itself (also see data recovery). I bought a 300 gig HDD last December and it came with a 5 year warranty, OTOH most modern USB flash drives have 1-2 year warranties...I do wonder why. Also the reason why your claim could be remotely true is data processing volume: A HDD while in active use can process tens to even hundreds of gigs worth of reads and writes every day for years, while Flash based solutions usually only need to handle a minuscule portion of that traffic volume during its lifetime. -G3, 13:49, 26 April 2007 (UTC)

Nah, GP is correct about unreported bad-sector marking. Isn't this productive. 128.232.248.113 21:39, 27 April 2007 (UTC)

It's a rather well known and well accepted fact that modern hard drives transparently reallocated bad sectors to spare sectors on the disk. However, I've never heard of anything even close to 20% spare sectors and for a large number of reasons, I highly doubt it's true. The number is likely much closer to 1% or less. In any case, as they say, extraordinary claims require extraordinary proof of which there is none from what i can see Nil Einne 18:15, 2 December 2007 (UTC)

Date errors in Cite Web refs

I just read this article and noticed three "cite web" references that claim the sources were retrieved on May 25th 2007, which is presently 26 days in the future. Must have been off by one month: http://en.wiki.x.io/w/index.php?title=Hybrid_drive&diff=prev&oldid=126016993 I'll just fix the month number in the three cites, for tidiness sake.Birdbrainscan 17:37, 30 April 2007 (UTC) Okay, I fixed them. It was two, not three cases. The diff link shows the edits were actually on April 25, by Verbatim9. Anyway thanks to all contributors for an interesting page. I'm sure this one will continue to develop rapidly along with its subject ;-) Birdbrainscan 17:40, 30 April 2007 (UTC)

conflicting statements on hard drive wear

Under Benefits / Improved Reliability, we read: "As the platters won't be spinning nearly as much, the wear and tear on the hard drive is drastically reduced."

But under Drawbacks / Increased frequency of spin-up, we read: "A hard drive, once spinning, suffers almost no wear."

So which is less stressful on the drive? More spinups/downs and less spinning, or fewer spinups/downs and more spinning?

Jbmweb1 (talk) 21:04, 21 January 2008 (UTC)

Spin-up and spin-down have classically always been associated with a lot of wear on hard drives [1]. During those operations, the chance of a head crash is much greater. Some even go as far as saying that the life of a hard drive is defined (or restricted) by the number of spin-ups it can do. Notice how there are no sources in the "advantages" section: there are simply no reliable sources (AFAIK) for the claims made by manufacturers about these drives. Theoretically, it should give a small performance boost on low-memory systems and when resuming from hibernation, but various magazines have debunked claims of any speed increase. Increased reliability is, to say the least, very doubtful, even when taking external influences into account. Remains is the decreased power usage; spinning down hard drives has always been used as a way to reduce the power usage of a hard drive (at the cost of reliability, and this also works with regular drives). This article needs work; I'm retagging it {{NPOV}} and {{Not verified}}. – Chip Zero 18:02, 22 January 2008 (UTC)

what this may actually be good for (but I don't KNOW)

Most computers I know (running windows) will hit the drive every couple of seconds with some very short access. I don't really know what happens under the hood during these, but I doubt any large amount of data is transferred. And because it is so regular, I doubt it reads/writes random data. Maybe these small disk access operations could be avoided. Even an autosave of documents or files could go into flash. As opposed to RAM, the data written to flash is persisted and will survive power failure, which is very important. I would assume that the drive will do some journaling so that it knows what tasks to complete as soon as power is back on.

I think those who see hybrid drive as an extention to RAM are missing the point. Hybrid drive will not help you in video editing and it is not intended as an extention or replacement of RAM. I believe it could be helpful in day-to-day office scenarios. The more RAM you have, the more can be cached in RAM and the less disk access you will need. Ideally, everything is cached in RAM and the HDD's flash memory is used exclusively to cache write operations. In that scenario running office applications I assume the HDD will not be spun up very often.

Platinum HDD

Is the Platinum HDD a standard, or a brand name? In the latter case, it looks like the last section might be an advertisement. —Preceding unsigned comment added by Tdvance (talk • contribs) 19:32, 27 September 2008 (UTC)

As far as I know Platinum HDD is a product, an example of a Hybrid RAM drive. There was a page about it, which did not seem to be warranted (no references to establish notability, etc.) It definitely needs cleanup (grammar, etc.). I merged it in here in case any of it was useful to expand coverage of Hybrid RAM drive. Zodon (talk) 05:48, 29 September 2008 (UTC)

Platinum HDD is a product, not a technology; there is nothing notably significant about it. I don't see how it qualifies as inclusion in this article. --A D Monroe III (talk) 00:42, 5 November 2008 (UTC)

As noted, the material was merged in here (a quicky dump it all in type merge). It definitely needs cleanup. I didn't just toss it because having specific examples and references can help understanding something. (i.e. it might serve as nucleus to expand details of hybrid ram drives (less specifics about that device more about the class), or use as an example of such) Since these are not common devices, it can help to have a place to start. Zodon (talk) 01:22, 5 November 2008 (UTC)

Software/dubious

The section which suggests this can be implemented in software is misleading [2]. Implementing it in system RAM misses the key advantage of non-votality (unless you have a battery backup/UPS). Ready Boast, while using flash memory (not system RAM as the article seems to suggest), is quite different from what is described here. Amongst other things, it's designed so the flash drive can be removed at any time without the system suffering. Nil Einne (talk) 14:28, 2 January 2009 (UTC)

---

I took a stab at rewriting the paragraph in question. It looked like too many ideas got mixed together there (Ready Boost implementing a hybrid flash drive; disk caching making a hybrid RAM drive). I tried to separate them out, but left the dubious tag in in case didn't adequately address the problem (if you think it adequately clear now, we can remove the tag).

I don't know why SuperCache II is specifically mentioned there. A little bit of research suggests it is just another disk cache program. Probably should be replaced with a general link to disk caches, unfortunately I haven't found much coverage of disk caching on Wikipedia. (What I have found seems mostly tied up with virtual memory, rather than dealing with file caches, etc.) Zodon (talk) 00:05, 3 January 2009 (UTC)

---

Hahahaaa, Nil Einne! Are you kidding? You are killing your "argument" in second first sentence: “(unless you have a battery backup/UPS)“.

Exactly! That’s the point! And UPS is better than battery backup, because it lets you write it to HDD, instead of just hoping that power will come back before the battery dies.

I have implemented such a thing myself under Linux. As opposed to Windows people, who never actually use their system, but just click on colorful software from others, this is actually quite easy to us:

Buy a UPS, add a bar of RAM, configure and load the driver... DONE!

The only one dubious is your ability to imagine things outside of your tiny world!

— 88.77.140.123 (talk) 08:31, 5 September 2009 (UTC)

---