Slashdot Mirror
Samsung's Hybrid Hard Drive Exposed
Erica Campbell writes "Samsung is preparing to release a new
Flash memory-assisted computer hard drive
that boasts improved performance, reduced energy consumption, a faster boot time, and better reliability. The new hybrid hard drive will be released around the same time as the upcoming Windows Vista operating system and will be one of the first hardware designed specifically to benefit from it."
that buffer is fucking huge. Laptops awesome, wonder when they'll actually work on a regular size one though. Then again, seeing as it's gonna be the first batch out the door, potential issues from what is practically a new drive type will scare me, and my wallet away.
Wonderful idea for the manufacturers, flash drives only get so many read/write cycles before they go T.U. Not so good for the consumers.
Looks like Samsung and Microsoft designed this together.s /HardDiskDrive_20050425_0000117556.htm
http://www.samsung.com/Products/HardDiskDrive/new
It was on display at WinHEC in April 2005.
Sorry, don't know how to link to one of the Caches, but here is the text of the article:
Samsung's HHD prototype
Samsung is preparing to release a new Flash memory-assisted computer hard drive that boasts improved performance, reduced energy consumption, a faster boot time, and better reliability. The new hybrid hard drive will be released around the same time as the upcoming Windows Vista operating system and will be one of the first hardware designed specifically to benefit from it.
Samsung's HHD - faster boot and resume on Vista
In mid-May 2006, Samsung unveiled a prototype hybrid hard drive (HHD) at WinHEC, the Windows Hardware Engineering Conference. Samsung's prototype HHDs have a buffer of 128 or 256 MB, much larger than the 8-16 MB of cache in current hard drives. This new buffer differs from the existing cache buffer on hard drives not only in size but also in structure, composition, and qualities. Conventional cache is made out of volatile memory that is erased when the drive is powered down. HHDs add another layer of cache consisting of Flash memory that is non-volatile and can be accessed quickly when the drive is powered on. Adding a large buffer to a hard drive can also reduce the drive's power consumption, thereby increasing the battery life, and reducing the time required for the system to resume its operation after suspension. Indeed, boot or resume time will occur about twice as fast as conventional hard disk drives, saving 8-25 seconds, and laptop batteries will provide 20 - 30 minutes more power. Another added bonus of the HHD is the improved reliability due to less mechanical wear and tear.
Samsung and other manufacturers are currently pursuing Solid State Drive (SSD) technology (to be covered in an upcoming TFOT article). Currently Flash prices are too high to allow SSDs to replace standard hard drives of any reasonable size and, although Flash prices are continually falling, it will be several years until such a drive will become affordable to most users. Here enters the near-term solution for enjoying improved performance at a reasonable price - the hybrid hard drive, combining the low cost and large storage capacity of conventional hard drive technology with quick and low-power Flash memory.
Apart from the reduction in Flash memory prices, hard drive manufacturers such as Samsung believe that we are about to undergo a major storage revolution in the next few years due to the upcoming release of Windows Vista. This new operating system from Microsoft will introduce three new performance-enhancing technologies: SuperFetch, ReadyBoost, and ReadyDrive. According to Microsoft, "SuperFetch understands which applications you use most, and preloads these applications into memory, so your system is more responsive". Windows ReadyBoost allows users to use a removable Flash memory device such as a USB thumb drive to improve system performance. ReadyBoost retrieves data stored on the Flash memory more quickly than data stored on the hard disk, decreasing the interval until the PC responds. Windows ReadyDrive enables Vista-based PCs equipped with an HHD to boot up faster, resume from hibernate in less time, preserve battery power, and improve hard disk reliability.
Hard drive platters won't have to spin as much
Hard disk platters are components of hard disk drives that consist of circular rigid disks that store magnetic data. While the platters in conventional hard drives rotate most of the time, thereby consuming a great deal of power, the platters in HHDs are usually at rest, as if they were off. In HHDs, incoming data is generally written to the Flash buffer and any saved documents are saved to the buffer, instead of being written to the hard drive each time. Only when the Flash buffer is almost full or when the user accesses a new file that is not stored on the buffer, will the HHD platter rotate or "spin up". Thus, the battery power of laptops with HHDs is preserved, extending battery life.
To learn more about Samsung's hybrid hard drive technology, TF
What's so different about Vista that makes this drive benefit from Vista. Will the drive not work in Windows XP, Linux or Mac OSX machines?
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
[nitpick]I know you're just trying to be funny, but the site was suspended due to exceeding its CPU quota, not its memory quota.[/nitpick]
Given Apple's strong relationship with Samsung (iPod shuffle+nano memory both come from Samsung, I believe- and I'm almost positive Samsung has supplied RAM to apple on+off since the golden olden days), what do others think about the possibility of this ending up in a Powerbook, er, Macbook Pro- and 10.5 being designed to take advantage of it?
Apple can be hit or miss with the latest and greatest- they took forever with USB2 (yeah yeah, firewire blah blah) and lagged behind a lot of the smaller laptop mafacturers with Expresscard (given there's next to nothing for expresscard, who can blame them?)...it'll be interesting to see if Apple thinks this is a win or lose technology...
Please help metamoderate.
I think the cache is designed to help with booting and suspend/restore, so it shouldn't be written too much. With a large enough flash buffer, it should be able to least for the normal life of the drive.
The number of read/write cycles is now typically sufficient to write at full speed 24/7 for 3-4 years.
"Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
"It's a tarp!" -- Dyslexic Admiral Ackbar
You're mentioning aged technology. Flash mems have improved since then, plus, it's slightly different technology.
Additionally, do you honestly think any company (Intel, Microsoft, Samsung) would back this technology if it was limited to R/W cycles in thousands?
Last but not least, such hard drives will also store data which stays more consistent than regular data. It could store vital boot files, files to your most common applications, etcetera. In other words, files that do not change much over time. It's not like you're going to save your most frequently used documents to this section of the drive.
So to sum things up, you will not have to worry about the SSD part of the drive. It will probably even outlast the mechanical part of the drive.
Full Tilt
You can have a similar effect now by using a flash drive as your root partition, or as a swap partition. Keep in mind that using it as a swap partition would make the drive age faster.
-jX
Don't you just love politics? It's like a comedy of errors.
Flash technology seems promising and looks poised to take over devices that would be better off using solid state components (laptops, etc) that traditionally don't. I've wanted to invest in Samsung and flash technology in general. Samsung seems to only be on the Asian markets, is this so? Does anyone know of and good mutual funds/ETFs that allows one to invest in this specific tech sector?
why run from Vincenzo?
Why don't they just flat out say they don't know when it's going to be released?
Why would you want your RAM to be unused? Unused RAM is useless RAM. Seriously.
I'm sure that Vista is smart enough to free up the RAM that SuperFetch is using if it could be better used for something else. It's really nothing more than a more pro-active version of the disc-cacheing that every operating system already uses.
If Vista knows about the CF, why does it need to be on the hard disk itself? It sounds like all the heavy lifting is being done by Vista anyways. WOuldn't it make more sense just to use any CF attached to the system for this caching, etc, and use normal hard disks instead? That way adding CF to a PC would improve its performance, no matter what type of hard disks you have attached.
Captain Cold takes over the world.
What happens when the flash dies?
The same thing that happens when your hard drive crashes - kiss the data goobye. That's what backups are for. You DO backup every day, don't you?
Seven puppies were harmed during the making of this post.
for how long you can chain a bunch of fat Linux programmers to your house with Mountain Dews (with caffeine) and french fries.
Forever, then?
Seven puppies were harmed during the making of this post.
the site was suspended due to exceeding its CPU quota, not its memory quota.
So maybe if they had preloaded the site into memory, we wouldn't have this problem...
Seven puppies were harmed during the making of this post.
Additionally, do you honestly think any company (Intel, Microsoft, Samsung) would back this technology if it was limited to R/W cycles in thousands?
Oh? Last time I checked...my xp seems to stop working after only several hundered read/writes, funny that.
We wrote a driver to read and write fat 16 flash drives for an embeded system. The testing for it wrote and read full speed 24/7 for two weeks before they died. I assumed that was because of the limited read write settings. Or is it possible the low quality connection was to blame? Doesn't really matter they were only used to transfer settings. As any one whos had to support them knows, they often just die for no apparent reason. I'm not convinced that this is a system I'd want my data on.
Well.. maybe. Or Maybe not. But Definitely not sort of.
What would be neat is if you could swap out flash drives in the event of a failure. Or upgrade the flash drive capacity. I'd be more interested in that than a permanently integrated flash drive. You're correct to be skeptical of its lifespan.
Well then, good news for you: Vista supports a feature called ReadyBoost, which can use just about any flash memory device (e.g. a cheap USB thumb drive) as a cache to improve performance.
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
Not true. If you write to ALL of the writable adressable area of a flashram , you will not get over 200,000 full writes on average despite the lies. In fact, the parts fail in 2 or 3 weeks of lab benchtests.
The flash fanatics keep modding down these facts to -1 for some insane reason here.
Flash has LIMITED write life.
The devices spread the data around to hide the limited write cycle life, and uses error correction to hide the limited write cycle life.
At some point its worthless.
Flash is idiotic for a backing store (virtual memory) based hard drive. And atomic-commit algorithms and other safety mechanisms for structure preservation and corruption avoidance such as "Journaling" only make the chatter worse.
All the disk chatter destroys the lifespan of the flash part.
Worse... flash is SLOW for lots of non-paralell-capable individual 512 byte requests, which typically are not spread across multiple flash parts.
True, a megabyte read can be fast in flash, but lots of random 512 byte reads or writes are far slower than a modern hard drive STILL in 2006. (15,000 rpm scsi from 7 diff manufacturers for example).
But the article is about hard drives... still.. its hopeless and foolish.
people who use their computers a lot will have data corruption earlier... all due to flash problems
How is this an improvement?
Because the _latency_ of flash is dramatically lower than mechanical hard disks.
You are looking at throughput, not latency, which is _vastly_ more important when talking about average access patterns. This is why an older SCSI drive with markedly lower throughput, but significantly better latency, will often perform better (especially for things like swap).
(Not to mention, your estimate of a 7200rpm drive is pretty generous to the tune of nearly 2x real-life performance).
I understand that there are other factors in play when accessing the hard disk, but.. I digress. Is this supposed to be a cheap way for Joe Schmoe to upgrade performance?
Yes. More accurately, cheap *and easy*.
"Don't buy 1GB of RAM for $100, but a 1GB flash drive for $30 and get 1/109th of the performance upgrade!!"
Firstly, it's going to deliver a significantly better benefit than than.
Secondly, upgrading RAM requires opening the case and putting it in. Most people are not comfortable with opening the case in the first place, let alone mucking around inside the thing possibly breaking stuff.
Modern flash is in the 10s of millions now; the flash in your $20 thumb drive is just old trickle down tech in the 100k write range. 30 million write cycles should last about the expected lifespan of the computer - 3 to 4 years.
moox. for a new generation.
Heck, MS says XP is "more secure". More secure than what?
Than previous editions of Windows, of course, and they're right.
I just had to clean my wife's laptop that is SP2 and fully patched with MS Windows Defender, MS Windows firewall and AVG anti virus and the thing has spyware crap on it that was bringing it to its knees.
And are Defender and AVG kept up to date? Is she running as an admin and installing any old crap she comes across? Is the firewall actually running?
A single anecdote proves nothing; I can attest to three XP machines that I personally use that are perfectly clean and have been for serveral years. Before blaming the OS, I'd check with your wife about how she was actually using the machine.
Yes, I do think any one of those companies would back any technology if that technology would make them a profit.
I can see that from Samsung, but neither Intel nor MS are going to be producing or selling these things, nor any hardware or software that relies on them. They're not going to stand to make any money on them, but will take a knock to their reputations if they back them and they're crap. Perhaps MS won't care, but Intel has serious competition from AMD, and can't be quite that cavalier.
It's official. Most of you are morons.
Out of curiosity, how similar is Vista's ReadyBoost feature to just mounting a USB drive as swap in Linux?
/proc/sys/vm/swappiness ?
Not really similar at all.
Can you hot-unplug the drive in Linux if it's being used for swap?
Nope, your system will crash unless you swapoff first (and of course that will fail if you're using more memory than you have physical RAM).
According to a FAQ on ReadyBoost I found, Vista will back up the pagefile to disk so it's not a catastrophe if you yank out the USB stick.
Correct. The data on the USB stick is used as a cache, not swap.
And is there any setting in Linux to tweak to let the system know you've got a fast swap partition, other than simply monkeying with
If there were, how would you want this information to affect Linux's behavior?
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
Except that Linux does preemptive swapping long before you run out of RAM, so that if something suddenly needs it, there's no huge delay whilst things get swapped out.
We did some basic research with Flash / HDD hybrids two years ago. As such disks weren't available, yet, we were using a real (Notebook)HDD and a IDE-Flash-HDD in parallel.
." in the root of a freshly booted system).
Our goal was to minimize energy consumption for mobile devices (i.e. not a lot of ram available for caching and the device is switched off repeatedly to save energy).
Using a very sensitive (time resolution wise) energy measurement device, we determined, that most energy was consumed by moving the heads into position. The difference was substancial: Around 0.63W for the HDD spinning idle and about 5.3W during heavy seeking (e.g. trigered by a "find
We decided to not use the flash as cache (flash is quick to read, but slow to write) and just put the relatively static metadata (directory structure, inode tables...) onto the Flash drive, but keep there files and data on the HDD, as each directory access triggered a expensive seek, but delivered very few data, compared to reading a file.
To simulate our mobile device we used a Linux-System limited to 32 ram to prevent the system from excessive caching.
We observed up to a factor 8 reduced energy consumption and as a surprising side effect a factor 6 increase in speed!
When increasing the available Ram, this advantage quickly vanished on repated benchmark runs, as the System appearently cached the directory structure very effectively. The first run after booting however still performed substancially better with our system, no matter the amout of ram. (And this was our target useage profile: Power on, search something, Power off).
As the code was an embarrassingly ugly hack to the ext2 driver and we envisioned trouble keeping the hdd with the data and the flash-hdd in sync, it was not persued further.
However with hybrid drives becoming available, it might be worth a more detailed analysis...