Samsung Mass Produces Fast 256GB SSDs

Lucas123 writes "Samsung said it's now mass producing a 256GB solid state disk that it says has sequential read/write rates of 220MB/sec and 200/MBsec, respectively. Samsung said it focused on narrowing the disparity of read/write rates on its SSD drive with this model by interleaving NAND flash chips using eight channels, the same way Intel boosts its X25 SSD. The drive doubles the performance of Samsung's previous 64GB and 128GB SSDs. 'The 256GB SSD launches applications 10 times faster than the fastest 7200rpm notebook HDD,' Samsung said in a statement."

Comments

cant wait

[ionix5891]

to put some of these into my servers

Re:cant wait

[negRo_slim]

to put some of these into my servers

ditto, but im waiting for permanent data erasure to become a little more mature. i understand the wear leveling incorporated into SSD can cause current programs to stumble.

Re:cant wait

[timeOday]

Me neither. We spent weeks (which translates to tens of thousands of dollars) benchmarking and optimizing a database app. The thought of accelerating it by a factor of 5-10x with a simple hardware upgrade is stunning.

Re:cant wait

[Shard013]

ditto, but im waiting for permanent data erasure to become a little more mature. i understand the wear leveling incorporated into SSD can cause current programs to stumble.

The mean time to fail in the new SSD's is a bucket load better that most regular HDD's.

http://www.google.com/search?q=ssd+mean+failure+vs+hdd

Between the 10,000-100,000 writes and the logic used to try not to rewrite the same place over and over they do quite well.

Re:cant wait

[cheater512]

One problem is they write in blocks (128kb?) and the entire block needs to be erased to modify a single byte.

That chews those writes faster than you'd expect, and write leveling can only do so much especially if the drive is mostly full.

Re:cant wait

[Zymergy]

Here is a Mil-Spec approved permanent SSD erasure device: http://static.howstuffworks.com/gif/hammer-1.jpg

Re:cant wait

[Anonymous Coward]

No, DB's arent happy with SSD. DB's are slower on SSD.

Re:cant wait

[magarity]

We spent weeks benchmarking and optimizing a database
 
And just think of all the people who have crappy unoptimized databases who can just buy these and leave their design as-is!

Re:cant wait

[Free the Cowards]

I could be wrong, but you sound like you're being sarcastic, which is a pretty stupid attitude to have here.

Let's say you have a crappy unoptimized database. You can spend tens of thousands of dollars' worth of programmer time to fix it up and optimize it so that it runs fast on your current hardware. Or you can spend perhaps one tenth of the money to upgrade to a super-fast disk, achieving the same end result. Which one is the smarter move?

Re:cant wait

[mrobinso]

Fixing the database is the right move. If it costs 10's of thousands of programmer dollars to fix it I'd say you have far bigger problems than an unoptimized database. I mean, programmers should only have access to selects updates and deletes. If your programmers are designing your databases, no wonder they're borked.

On topic, I can wait for the SSDs. The price of the 250s, mass produced or no, still presents as a problem for a business case to run out and load up on these. Hardly stocking stuffers.

Re:cant wait

[b1t r0t]

Here is an even more permanent erasure method.

Re:cant wait

[Anonymous Coward]

You won't be putting this in your db server, I guarantee it.

Re:cant wait

[magarity]

could be wrong, but you sound like you're being sarcastic
 
I was being 50% sarcastic - some people get lucky and a technological advance allows them to fix poorly performing design for the time being (they then turn into) Other people assume that some advance will come along and then either it doesn't or they can't afford it. Finally, there are people who see that a poorly performing design is going to catch them up and they proactively fix it before anyone loses their job over the order taking system crashing for a couple of days due to heavy load.
 
Databases are frequently the workhorse backend on which the business runs; deciding whether to spend $$$ now to fix poor initial design is like deciding whether to put safety bladders in the gas tanks of Pintos.

Re:cant wait

[Zero__Kelvin]

1. Writes are byte / word size

2. Reads are byte word size

3. Erasures are in quantums of BLOCKSIZE
4. BLOCKSIZE on a typical NAND device can be 2k, or 4k, or some other significantly smaller value than 128k

5. The caveat is that it is necessary to erase the whole block to write a byte when a bit needs to be changed from its erase state. e.g. :
   
   If the erase state is 0 and a bit needs to be cleared (it holds a 1 in or scenario, and we want it to be zero) then it is necessary to erase the whole block. This (obviously) means copying the block contents to RAM, zeroing the FLASH page in which the byte resides, and then writing the page back to FLASH. It sounds worse than it is, and ultimately the overhead doesn't put a dent in the difference between using spinning media and FLASH. For example, what is the overhead to change a single byte on a hard disk?

   1. Spin up the platter if needed
   2. Seek to the filesystem metadata and find out where the file resides.
   3. Read in that data, then use it to determine where on the disk it is
   4. Seek to the location where the file resides
   5. Read the sector that needs to change
   6. Change the data in RAM
   7. Write the sector to the platter

   8. Update the metadata (itemization of steps not included; you get the idea.)

Re:cant wait

[Anonymous Coward]

im waiting for permanent data erasure to become a little more mature.

SSD, meet furnace.

Re:cant wait

[Burning1]

Who told you this?

Wear leveling is handled at the firmware level, and should be transparent to the application. There's no reason it should cause issues any more than RAID, error correction or sector re-mapping causes issues on current hard drives.

Re:cant wait

[Burning1]

How often do you edit a single byte without updating the rest of the block? How often do you write period? And, what difference does it make weather you burned out a single byte within a block, or the entire block?

Furthermore, wear leveling should keep this from burning out a specific block on the device: every time the block is re-written, it goes on a different part of the disk.

Re:cant wait

[Free the Cowards]

Well, the guy who brought up database optimization said that "we", which implies at least two people, spent "weeks" optimizing a database which was too slow on spinning media. Multiple people with the appropriate expertise spending that much time is going to cost many tens of thousands of dollars.

Perhaps they had "far bigger problems". Didn't sound like it though.

Re:cant wait

[cheater512]

It seems (to me anyway) that people dont think a single byte being defective is a big deal, when in fact its dead blocks which are somewhat larger.

Logs, most databases and many other things have frequent small writes.

It means that the SSD can be fried somewhat faster than people think.
Wear leveling just means it all fails at the same time instead of some bits failing sooner than others.

Re:cant wait

[Henkc]

Not everyone works on simple 5-table MySQL DBs for their online pr0n collection. I know it must be hard to imagine, but some people actually work on multi-million dollar systems which are distributed across multi-node clusters, with TBs of data, etc. Some of those people actually lurk around here too...

Some of us can't wait for SSDs either - like the poster said: to be able to instantly gain 5-10x performance is staggering and worth every cent. When last did computing tech provide that kind of performance jump?

Re:cant wait

[fractoid]

Outside consulting contractor: $120/hr
One week's onsite work: 38 hours.

Mathemagic says that hiring two consultants for more than a week each would be above $10k.

Re:cant wait

[marcosdumay]

Oh, yes, you have the option of spending 100k beforehand and fix that O(e^n) algorithm, or, alternatively, spend some 30k upgrading your disks and waiting a few more months before more data is added and the app crashes again, and, then you spend those 100k.

That is quite a big interest rate here.

Re:cant wait

[MrZilla]

That chews those writes faster than you'd expect, and write leveling can only do so much especially if the drive is mostly full.

Not really. With dynamic wear leveling, you will use the entire disk, not only the empty parts. Some SSD manufacturers (and their customers) have found out the hard way why only using the empty parts is a bad idea...

Re:cant wait

[Free the Cowards]

If the speed increase due to optimization is the same as the speed increase due to a hardware upgrade, then the benefits are the same....

I don't know why everyone thinks that a hardware upgrade's benefits will be wiped out nearly immediately but carefully optimizing the software will last forever. It just makes no sense!

I can only conclude that Slashdot is home to a cult of optimization worshippers.

Re:cant wait

[stubob]

Do they seriously have a howstuffworks page for a hammer?

Re:cant wait

[marcosdumay]

"I don't know why everyone thinks that a hardware upgrade's benefits will be wiped out nearly immediately but carefully optimizing the software will last forever."

Because too often a slow program is slow because it uses a O(n^8) algorithm where it should be using a O(n) one. So, every doubling at the hardware speed after optimization will lead to a doubling on the amount of data your program can deal with, but every 256 times hardwre spee-up will lead to a doubling on the amount of data the program can handle before optimization.

After some point, even Moore's law get slow.

Re:cant wait

[Anonymous Coward]

Rofl rofl rofl that has to be the funniest way to promote a blender ever. Thanks.

Yeah but it costs how much?

[jeepmeister]

My left nut is an unreasonable price.

Re:Yeah but it costs how much?

[drpimp]

My left nut is reserved for a hot female if that is what it costs to court her. So the question is ...

A) my right nut has plenty of semen to procreate and force her to be a part of my life for at least 18 years.
B) my right nut can be used to buy the SSD.

Shit... decisions decisions! What happens whent he next fastest SSD comes out, I have no nuts left. FUCK!

Re:Yeah but it costs how much?

[Anonymous Coward]

Why? You're not using it anyway.

Re:Yeah but it costs how much?

[v1]

My left nut is an unreasonable price.

"don't worry, son, that's why God gave you two."

Re:Yeah but it costs how much?

[Anonymous Coward]

My left nut is an unreasonable price.

"don't worry, son, that's why God gave you two."

Two left nuts? Isn't that the cure for two left feet?

Re:Yeah but it costs how much?

[HTH NE1]

Yeah, it'll be 1 to 3 organs, depending on demand. Demand for soft organs is red-hot right now, so it's pretty well sure to be a triple-donation. Probably the usual combo: eyeball, kidney, testicle. If they want bone, they'll take an arm or a leg, but we haven't done a limb-cut in days.

Re:Yeah but it costs how much?

[Toe, The]

For comparison, an Imation 128GB costs between $1,900 and $2,500.

I'd guess that the Samsung 256GB ones would therefore cost in the range of $5,000 +/- $2,000 (probably more likely +, given there's usually a premium for new and for bigger).

Re:Yeah but it costs how much?

[Anonymous Coward]

No, he did not. Only one. :(

Re:Yeah but it costs how much?

[BKX]

You've been watching Lexx lately haven't you. Next thing you'll know, you'll start referring to His Great Shadow every third sentence.

Re:Yeah but it costs how much?

[Lifthrasir]

His Divine Shadow, but who's counting?

Re:Yeah but it costs how much?

[Anonymous Coward]

People have two left nuts? That's news to me...

Re:Yeah but it costs how much?

[Anonymous Coward]

Perhaps your more inclined to pass over your right.

Re:Yeah but it costs how much?

[Anonymous Coward]

troll ?? wtf. Someone has no sense of humor or is a totally n00b and doesn't know what a troll is!

Re:Yeah but it costs how much?

[Anonymous Coward]

Awesome. "The Big Red One." Too bad I can't find that clip online anywhere.

Well, this one is not bad:
http://www.youtube.com/watch?v=zWyBrUvXLcc

Re:Yeah but it costs how much?

[magarity]

Good god, man, only suckers with corporate accounts have to shop at CDW. $1,900 for an MLC drive? Try $300 for the same speed specs. If you want SLC you'll have to double that, but still, CDW is way overpriced these days.

Re:Yeah but it costs how much?

[Toe, The]

It's a baseline. :)

And, as one who used to buy from CDW on a corporate account... they give a discount for those accounts. Though I have to admit it's usually rather pathetic. For big-ticket items, I would usually have to talk them down or shop elsewhere.

Re:Yeah but it costs how much?

[tomzyk]

My left nut is an unreasonable price.

"don't worry, son, that's why God gave you two."

Two left nuts? Isn't that the cure for two left feet?

I thought the cure for two left feet was something along the lines of taking dancing lessons.
Therefore, the cure for two left nuts would be... hookers?

Re:Yeah but it costs how much?

[halcyon1234]

My left nut is an unreasonable price.

"don't worry, son, that's why God gave you two."

Except that last year he wanted an iPhone...

10,000 RPM

[neoform]

Does this mean it's no longer useful to buy 10,000RPM drives? 10x faster? Sweet.

Re:10,000 RPM

[TooMuchToDo]

Just imagine the power savings as well. Also, they should last an order of a magnitude longer than media that needs to spin all the time.

As soon as these get cheaper and have more capacity, spinning media is dead.

Re:10,000 RPM

[bunratty]

Spinning media already is dead. It's just that no one's told it yet.

Actually, spinning media will continue to be used in servers that need huge capacities of storage. But for cheaper devices, the speed, energy efficiency, durability, and price of solid state drives will effectively make using spinning media obsolete in the next few years.

Re:10,000 RPM

[Archangel Michael]

It isn't dead until Netcraft confirms it.

Re:10,000 RPM

[CopaceticOpus]

1 TB of SSD today = 17 * $150 = $2,550.

1 TB conventional storage = $95.

SSDs are still over 25 times as expensive. They will improve quickly, but they need to hit a moving target to kill conventional drives.

Re:10,000 RPM

[bunratty]

SSDs are still over 25 times as expensive for 1 TB of storage. Fixed that for you.

64 GB SSD today = $150.
80 GB hard disk = $40.
If you need only 64 GB of storage, as most handhelds, laptops, and desktops do, SSDs are only about four times more expensive today. You can expect SSDs to become cheaper than hard disks in about two years, at least for the smaller capacity drives.

Re:10,000 RPM

[DogDude]

Where can you get a 64 GB SSD for $150? I'd buy about 10 of those right now if I could find 'em.

Re:10,000 RPM

[ignavus]

1 TB of SSD today = 17 * $150 = $2,550.

1 TB conventional storage = $95.

SSDs are still over 25 times as expensive. They will improve quickly, but they need to hit a moving target to kill conventional drives.

A perfect choice for RAVED - Redundant Array of Very Expensive Disks.

Re:10,000 RPM

[andy_t_roo]

your grand-parents post has a newegg link to a 60Gb SSD drive for 150 - its not quite the 64G that your parents post has, but its close enough (ie, you could probably find enough of a discount by shopping around to get to the same $/GB as the 150/64 you seem to want.)

Re:10,000 RPM

[i.of.the.storm]

Well at 80GB hard drive prices don't scale as well because hardly anyone makes 80GB hard drives anymore. Your comparison is a bit unfairly slanted, because while 64GB might have been fine 3 or 4 years ago, these days Windows takes up 11GB by itself, and in the age of HD movie downloading and double-digit GB games 64GB doesn't quite cut it. However, if you're going for more of an ultraportable type device you can probably get by with 64GB easily I suppose, and add on with an external HDD.

Re:10,000 RPM

[Teilo]

http://tinyurl.com/cheapssd

Probably nothing like a Seagate, but, technically they are SSD drives. I imagine they are probably more like big thumbdrives with a run-of-the-mill SATA controller slapped on them.

Re:10,000 RPM

[afidel]

Those are MLC flash, random writes are going to be horribly slow and will reduce the wear life to months in a server application. Would be perfectly usable in a netbook/laptop application though.

Re:10,000 RPM

[afidel]

Not expensive at all if what you need is IOPS instead of storage (though you would want SLC flash for a server app which is more expensive).

Re:10,000 RPM

[jarlsberg71]

Well if you're going to make analogies, keep them atleast the same form factor. Newegg has 2.5" 128GB SSD's, for $278. so 8 = $2224 and their 2.5" HDD? 300GB for $260. 4 = $1040 so twice the price, less power, and 3-10x faster

Re:10,000 RPM

[shermo]

Surely the cheap, huge capacity market is claimed by tape drives, and they're not going away any time soon. It'll be interesting to see if conventional hard drives can fill the middle niche, or if they get squeezed.

Re:10,000 RPM

[CopaceticOpus]

I agree that SSDs will find a niche soon, and that niche will continue to grow. My point is simply that spinning drives are nowhere near being a "dead" technology as the original post stated.

You can expect SSDs to become cheaper than hard disks in about two years, at least for the smaller capacity drives.

Is this just speculation? In any case, it seems likely. Every hard drive requires a calibrated motor and many other specialized parts. The cheapest hard drive can only be so cheap, and so at smaller sizes, SSDs may take over soon.

Re:10,000 RPM

[Anonymous Coward]

I prefer RAGED - Redundant Array of Greatly Expensive Disks.

Re:10,000 RPM

[kklein]

If you need only 64 GB of storage, as most handhelds, laptops, and desktops do

My laptop has a 250GB drive that's almost full with work files. I haven't had less than a few hundred GB of storage on a desktop for almost a decade.

Hell, even my iPod is 80GB, and almost full.

Are you from the past?

Re:10,000 RPM

[ogl_codemonkey]

This TinyURL redirects to:
http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&SubCategory=636&N=2013240636

Re:10,000 RPM

[Anonymous Coward]

Spinning Media isn't going anywhere. It will have a better price per GB for many years to come. And when I'm buying 2TB spinning drives for $150 a year or two from now, and $150 would get me a 64gb flash drive, the choice is going to be as clear then as it is now.

And when the 2TB flash drive costs $150, the 10TB spinning disk will also, and so on and so forth.

Re:10,000 RPM

[Anonymous Coward]

That's still really high end. I'm a university and almost nobody I know uses more than 60 GB. Gamers and movie downloaders are among those that make up the exception to the rule still.

Re:10,000 RPM

[khellendros1984]

I knew a large community of movie downloaders, so most of the people with whom I was familiar had several hundred gigabytes. I guess I'm just weird...

Re:10,000 RPM

[cheater512]

Calibrated motor? You think that they have to specially make motors which run at exactly 7,200rpm? O_o

They just use a feedback mechanism for precise positioning and speed.

Flash on the other hand chews up massive amounts of space on silicon.
That is quite expensive.

Re:10,000 RPM

[justinlee37]

The idea here is to run your applications on the SSD drive and all of your less-central media files (work documents, movies, game backups, etc.) on your HDD drives. That way the HDD only spins up when you are loading something, and if you have enough RAM, it will basically just spin up to load that data into the RAM for your applications to access (which are conveniently on the super-fast SSD drive).

Taking it from that approach, you could easily put your operating system on the SSD drive, along with a few games, keep your old harddrives, and call it good.

But yeah, the OP's insinuation that 64gb is enough "total" is ridiculous. I use 380gb on my home machine, and I still feel like I don't have enough room. I'm constantly forced to delete things.

At work we use almost 2TB to archive video files, along with another 1-2TB for backup.

Re:10,000 RPM

[Anonymous Coward]

64GB is peanuts. Maybe good for a cheap camera, or music player. My personal work-related files take up just a bit over 10TB plus another 20TB of backups. My non-work related files clock in at about 3TB. I will be a very long time until I switch to SSDs.

Re:10,000 RPM

[bunratty]

I searched for the lowest priced hard disk at newegg, and it turned out to be an 80 GB hard disk for $36. For most laptop and desktop computers, several tens of gigabytes of storage is plenty. When 64 GB SSDs get down to about $40, they should become the standard drives on low-end computers.

Re:10,000 RPM

[bunratty]

Where I work we use a file server to store files. Are you from the past?

Re:10,000 RPM

[vyruss000]

What about RAPED? That seems more appropriate...

(Reduntant Array of Patently Expensive Disks)

Re:10,000 RPM

[i.of.the.storm]

Sure, but my point was that the sweet spot in terms of price per gigabyte is a bit higher than 80GB as these days hard drives of that size aren't used much anymore - most computers you buy will have at least 160GB or so, and Zunes and iPods discontinued their 80GB models for the 120GB one. The economies of scale that are in favor of hard drives don't really work out at that point. And sure, 64GB SSDs will probably hit $40 in a year or so, but by then the equivalently priced hard drive will be 160GB or so. SSDs are advancing much faster than hard drives though, so at some point they will overtake hard drives in terms of price per gigabyte, but hard drive tech isn't completely stagnant and it's going to be a while yet before SSDs can compete on a similar price level.

Re:10,000 RPM

[Ant P.]

Nothing wrong with living in the past. I'm still living in 2007 with my 4GB laptop and I've never felt like I was short on free space. But then I don't really use that laptop for HD video editing or the like...

Re:10,000 RPM

[Eskarel]

Tape drives are for backups, and that's about it, because they're orders of magnitude slower than anything you'd ever want to randomly access data from, and seek times on a tape are what you might all horrendous.

Realistically they're not all that cheap either unless you're looking at seriously high volume data.

Re:10,000 RPM

[boot_img]

Don't think so

DLT Tapes ($25 for 40 GB + cost of drive) are now more expensive than hard drives.

They are even deader than spinning disk.

Re:10,000 RPM

[bunratty]

True. The backup system I bought last year is external hard disks connected to the file server by an eSATA interface. It's much faster and cheaper than tapes up to at least a few TBs. I suppose if you have many terabytes of data, or already have access to a good tape drive, tapes are still cheaper for backups.

Re:10,000 RPM

[shermo]

So, you're agreeing with me? They're the ultimate in high volume, cheap, slow storage.

To the cousin poster - would you propose that we do monthly-rotating daily backups of our small business server to spinning hard drives? Keep in mind that they need to be easily archivable and the cost of the media is more important than cost of the hardware.

Re:10,000 RPM

[nabsltd]

But why would anybody pay $36 for 80GB when they can pay $45 for 250GB?

Sure, you may not need more than 80GB right now, but when you get 170GB more for the price of a fast food meal, it's kind of a no-brainer. And, with the exception of 10K rpm drives, you pretty much have to buy a larger drive to get the latest, greatest technology.

Re:10,000 RPM

[nabsltd]

The idea here is to run your applications on the SSD drive and all of your less-central media files (work documents, movies, game backups, etc.) on your HDD drives.

That's another strike against SSDs.

Laptops generally don't have space for a second drive (and you sure don't want to lug an external drive around) unless you are content with the non-upgradeable, proprietary form-factor SSD drives that some laptops are offering now.

And, desktops have enough space that they can benefit from other technology options (>4GB RAM, RAID-10, etc.) that make SSDs less of a win. Seriously, for the price of a 256GB SSD that won't exhaust the write count too quickly, you can outfit a system with 16GB of RAM and four 300GB 10K drives in RAID-10.

Re:10,000 RPM

[pipingguy]

My replacement 60GB SSD (OCZ) was doing fine in my MacBook Pro, then I installed XP SP2 on the boot camp partition and Parallels. D'oh! Now I have to repartition and give XP more space for the Windows Updates. I haven't noticed much speed boost from the non-spinning drive, but maybe I've just gotten used to it.

Re:10,000 RPM

[Firethorn]

To the cousin poster - would you propose that we do monthly-rotating daily backups of our small business server to spinning hard drives? Keep in mind that they need to be easily archivable and the cost of the media is more important than cost of the hardware.

Are you doing off site? How many gigs are you backing up? You might find that a raid NAS array, possibly off site to be cheaper than you think - and fully capable of keeping all your backups. Or price a dozen or so external HDs.

So, you're agreeing with me? They're the ultimate in high volume, cheap, slow storage.

And hard drives will rule the world for a while when it comes to on-line, random access, but not requiring especially low latency.

Still, I find it interesting. Right now, on Newegg, the largest HD you can get is 1.5TB for $140. For $278 you can get a 128GB SSD. Call it $2/GB.

It wasn't that long ago that the HD was ~$300, and the SSD, $3k for a 80GB one. Matter of fact, Newegg still lists a 64GB model for $825. Back in 2004, a 250GB HD cost $250. Anyways - we're looking now at SSDs being available that are 'only' 1/12 the size of the largest consumer HD available at this time, for about double the price. Go back around a year, and you're looking at 10X the price for 1/12th the capacity. We went from a 120X disadvantage to a 24X advantage. That's a massive catchup, relatively speaking.

Keeping with Newegg - you can get a 120GB 2.5" drive for ~$50. So it's 5X as expensive to get the SSD - but the SSD is shockproof in comparison, and is demonstratively faster. Go large on the HD? 500GB for $110. Around a 12X disadvantage. At this rate I'll predict that SSDs will replace hard drives in laptops around 2010-2012. Leaning towards 2010. Shortly after that it'll take the server market, at least for systems that lean towards reads. 2016 or so for standard desktops.

Re:10,000 RPM

[Firethorn]

Is $152 close enough? G.SKILL FM-25S2S-64GB 2.5" 64GB SATA II Internal Solid state disk (SSD) - Retail

There's one available for $145, but it's write speed is 1/6th the slightly more expensive one, and it's read a quarter. Still MLC flash though, looking at the manufacturer's webpage. They have SLC - at 3.5 times the price. Reviews suggest the SLC is much, much faster. Advertised sequential writes/reads are a tad slower, but the random reads/writes aren't listed for the cheaper product.

Re:10,000 RPM

[techdojo]

Mechanical anything is pretty much toast. I'll look forward to having thumb-drive size movie rentals and a PC with no moving parts. Size, price, and noise will plummet together.

Off with their heads, says I.

_____________________________________
http://techdojo.org/

Re:10,000 RPM

[Firethorn]

One thing to remember is people like my parents and grandparents. I use over a TB of HD space. My parents haven't even used over 50 yet, and my grandparents even less.

Due to the mechanical components of hard drives, they aren't going to get much cheaper, even/especially in bulk - some of the bottom end we see are the manufacturers putting their old drives on fire sale.

Once the manufacturers can get a 40-80GB HD for LESS than they can get a hd*, I predict they'll start switching. They're already on the poor side for GB per $.

The cheapest laptop HD on newegg is $50 for 80GB(.625). Lots of 320GB($70,.22), 500GB($110,.22). The cheapest 3.5" HD is $36 for 80GB(.45). $42 for 160GB(.26). A 1TB one runs $95(.095), so per GB it's a much better deal, delivering 4 times the GB per dollar over the 'cheap' 80GB. The cheapest SSD is $20 for 4GB(5). Not very efficient, not even very fast. Going up - $145 for 64GB(2.27). $278 for 128GB(2.17) Not much economy of scale gained, but that's to be expected. 32GB seems to be where the SSDs start flattening out($84,32GB,2.63) at the moment.

Going by this - I figure 3 years before you start seriously seeing SSDs replacing hard drives in laptops - and it'll start on the low ends for cost savings, and the high end for performance.

It'll be another 5-10 before they start doing the same to desktops. Still, I figure upgrading will become common again - fast flash for OS and programs, cheap big HD for most multimedia.

*As long as it can be expected to last long enough that they don't have to do warranty work, and performs at least as well.

Re:10,000 RPM

[jbengt]

But why would anybody pay $36 for 80GB when they can pay $45 for 250GB?

An OEM who wants to list the low end notebook for $399 instead of $408

Re:10,000 RPM

[neokushan]

Wait wait wait...you ARE a University?

Re:10,000 RPM

[i.of.the.storm]

Hmm, nice calculations. I agree it's going to ake a while for flash to overtake SSD, and hopefully eventually we'll get dual/hybrid drives with maybe 32GB or 64GB of fast flash for OS and key programs and maybe 1TB for media and mass storage, which would be awesome. That is something I would love to have in a laptop, and hopefully in most cases the HDD wouldn't even need to spin up unless you're playing a movie or something, and you can even have it use the flash as a giant cache and the OS could copy the movie to the flash to take advantage of HDDs sequential access advantage and then not have to worry about the HDD for a while.

Re:10,000 RPM

[Eskarel]

No, I'm suggesting that what you're talking about isn't applicable to the discussion.
Tape drives and hard drives are not used for the same purpose. A discussion about solid state drives has absolutely nothing to do with tape drives.
No one is going to replace tape drives with SSD's, and no one is likely to start using HDD's for long term storage(at least not if they're serious and not a complete nutter).
This whole thread has been about how HDD's will survive in servers (due to large data volume) for a while, but that smaller drives in local machines will be replaced with SSD's rather quickly.

Re:10,000 RPM

[Anonymous Coward]

What an awful comparison! How did you got modded up. You're comparing 17 2.5" 60G SDD against a single 3.5" HDD?

Go ahead and buy 17 2.5" 60G HDDs to be fair, and that will cost you around $1000.

Next time you can compare it to one million of floppy disks.

Re:10,000 RPM

[Teilo]

Uh, yeah, and?

Re:10,000 RPM

[TooMuchToDo]

Silly techdojo, movie rentals will all be online a-la-Netflix. Physical media for rental is so 20th century.

Re:10,000 RPM

[TooMuchToDo]

Depends on the server app. Database? Definately. A web server that doesn't log? Good fit. It won't write to the drive that often, so the SSD should last fairly long. The drives would also be excellent for Google, as their Google File System is optimized to store data on their drives that isn't rewritten all that often (new data they search/index is added to their data collection, and usually doesn't overwrite the slightly older data). It's fairly close to Write Once, Read Many. More like Write Occasionally, Read Lots.

Re:10,000 RPM

[Jeremi]

Spinning media already is dead. It's just that no one's told it yet.

Talk like that will have it spinning in its grave.

Re:10,000 RPM

[Anonymous Coward]

OMG! It's full of naked people!

Re:10,000 RPM

[Anonymous Coward]

Whats this efficient energy requirement? No one cared about this back in the days but why wake up now? They say energy costs too much but thats only to blame for those who controll it.

Re:10,000 RPM

[icsx]

Why get 80GB when you can get 1TB for little over double of that price? Are you in selling/marketing department? They usually bend numbers...

Re:10,000 RPM

[muszek]

Google is your friend. From the meta description tag of netcraft.com:

"Netcraft provide monthly Internet research reports on the apparent death of FreeBSD."

Re:10,000 RPM

[Skye16]

That seems odd. I didn't know a single student in college who DIDN'T play games or download movies.

You must come from bizarro land where people go to university to pursue scholarly activities instead. Craziness.

Re:10,000 RPM

[eastlight_jim]

Your parents haven't even used over 50TB yet? That's appalling!

Re:10,000 RPM

[XMyth]

And....now people can click on the full URL without worrying about seeing a gaping anus?

Re:10,000 RPM

[Teilo]

Ah! Ok, then.

Re:10,000 RPM

[Anonymous Coward]

Just to be smarmy, it's actually RAED

Redundant Array of Expensive Disks

Re:10,000 RPM

[morgan_greywolf]

You can expect SSDs to become cheaper than hard disks in about two years, at least for the smaller capacity drives.

But .... that's what you SSD guys said 2 years ago. :-P

Re:10,000 RPM

[Bake]

Huh? What non-upgradable proprietary form-factor SSDs that some laptops have, are you talking about?

Please inform me, I'm finding it hard to believe that laptop manufacturers and SSD manufacturers are going out of their way to come up with a non-standard form-factor to fit hard drives into.

Re:10,000 RPM

[maxume]

As your storage per chip goes up, you can make smaller flash drives by using less chips. Hell, it is probably possible to use chips with major flaws (just have them present as a lower capacity chip).

Hard drives of any size need all of the same precision mechanical components (with some allowance for different numbers of platters), so the fixed costs on the low end will, at some point, give flash the advantage.

Re:10,000 RPM

[DarkAce911]

who would put in a DLT? LTO4 is the way to go, 800 GB uncompressed for $60 a tape.

Re:10,000 RPM

[richlv]

My laptop has a 250GB drive that's almost full with work files. I haven't had less than a few hundred GB of storage on a desktop for almost a decade.

that kinda depends on the work, doesn't it.
i have 40 or 80 gb hdd at the work (see, i don't even remember the exact size), and only recently i managed to fill it up (and then i removed some obsolete server restoration images which should really not reside on workstations anyway).
i know people who haven't filled up their 8gb drives, and a lot more who are using similar amounts of data on 40/80/200 gb drives.
once 64 gb ssd costs ~ the same as 60 gb hdd, market share will already have been swapped for those.

Re:10,000 RPM

[nabsltd]

There were several laptops where the SSD is either on the motherboard or otherwise non-standard. I don't know if they are still made.

Sorry, but I can't find where I read about them. Obviously, it was one of the larger manufacturers like Dell, Lenovo, etc.

Re:10,000 RPM

[Shinmizu]

But... snopes says Netcraft is just an urban myth.

Re:10,000 RPM

[nasor]

You can get 800 GB LTO4 tapes for about $65, still better than hard drives in bytes/dollar. In addition to being about 20-30% cheaper than hard drives in terms of space/dollar, no one will care if you drop the backup tape cartridge; the shock isn't going to scratch a platter or break some electronic part.

Re:10,000 RPM

[nasor]

True, but my 1 TB harddrive has been less than 1/3 full for months, and I don't really anticipate filling it any time in the next few years. I mean, I could copy 100 DVDs to the thing and still have hundreds of gigs worth of free. Assuming my storage needs don't grow enormously, I would probably rather have a SSD with a few hundred GB of space than a regular hard drive with TBs of space that I'm not likely to ever fill.

Re:10,000 RPM

[Firethorn]

50GB Smartass.

For that matter, over a patched windows install + office, I'd rate my GP's usage as 'insignificant'. Less than a GB between both sets. Mother's side used a bit more, as my grandmother ended up getting one of those fancy computer controlled sewing machines.

Mom&Dad? Mostly pictures, some documents. Might be over a GB, might not.

So I see a situation where computers come with 40-80GB of flash, and people who actually use storage space and don't simply stream from Itunes or whatever movie equivalents pop up, will simply pay $100-200 and slap in a big HD for their needs.

As far as it goes - flash has already killed anything smaller than 1.8" drives, and are busily killing the 1.8" ones. 2.5" will last a couple years longer, I figure as I said earlier. Another couple doublings in size and halvings of price, along with trickle down performance improvements and you'll be looking at many, many people buying them. Heck, for notebook/laptop makers, build the SSD into the motherboard.

Then, once they've started making serious inroads into laptops, you'll see HD development start to lag(writing on the wall), and SSD/Flash development continue to accelerate. We're seeing it now, after years of languishing in BIOSes and special purpose applications, now flash is in cell phones, PDAs, MP3 players, cameras, etc... And development is booming. They get to within a stone's throw of HD's and now they're competing with them as well.

The start of it all - the digital camera and mp3 player. They drove the development of flash.

Re:10,000 RPM

[Firethorn]

Going by the .22 for a 2.5" drive vs the 2.2 for a SSD, I'd say the SSD is 4 'generations' behind laptop drives. Thing is, SSDs are currently undergoing 2-3 intinerations/generations per HD intineration. So a SSD will double in capacity per $ every six months or so while it takes 12-18 months for the same with a HD.

So, my figuring is that it'll take 2 years for SSDs to catch up to where HDs are 'now', but the HDs will advance, adding another year before they catch up for real.

Re:10,000 RPM

[i.of.the.storm]

3 years sounds about right. Pretty cool stuff.

Re:10,000 RPM

[sbjornda]

no one will care if you drop the backup tape cartridge

One of our vendors offered a tape handling course. They claim that, with LTO, you can scramble bits on a tape if you drop it. I've never tried it myself.

--
.nosig

Re:10,000 RPM

[Firethorn]

SSD Price War Begins
SSDs To Match High-end Hard Disk Prices By 2010?

Looks like some others are predicting 2 years. It's interesting to note that in January of 2007 the price per GB for a SSD was $10. It was 32GB. Today, it's $2 and change for 128GB. That's two doublings of capacity, and two halvings of price, in under two years.

Re:10,000 RPM

[pjp6259]

If you're worried about a drive by goatsxing, you might want to turn on preview for tinyurl:
http://tinyurl.com/preview.php

Re:10,000 RPM

[iamhassi]

"Does this mean it's no longer useful to buy 10,000RPM drives?"

10k rpm? I'm more interested in replacing the 15k rpm drives.

Re:10,000 RPM

[iamhassi]

This same thing happened 5 years ago with 15k drives. A 73gb 15k SCSI drive was $400+, but a 80gb 7200rpm IDE drive was only $75. Now you can get that same 15,000 RPM drive for less than $50.

so just wait a few years. In 5 years that 256gb SSD will be $50, and $100 will buy 12tb.

Re:10,000 RPM

[CopaceticOpus]

So the question is this - in five years will most people find a 512GB SSD is adequate for their needs, or will they have found reasons to make use of the 12TB conventional drive? I would guess they will each have their uses.

Or on the other hand, in 5 years will SSDs be catching up to conventional drives? Murphy's Law seems likely to favor the pure silicon nature of SSDs over the electro-mechanical awkwardness of conventional drives.

Re:10,000 RPM

[steveg]

Define "almost a decade."

A decade ago I was pretty excited about 20G drives. I don't remember "a few hundred GB" available at any price for a desktop. We had A SAN that had over 3 TB for our servers (shared among all of them.) It was less than a million dollars -- but not a *lot* less.

We had a somewhat lower performance RAID array that had "only" 200G, and it was a lot cheaper. Less than $10,000. That wasn't used on a desktop either...

Re:10,000 RPM

[ogl_codemonkey]

... which only works for as long as you have the cookie it sets; and still leaves the delay of a pointless redirection, and of course the problem of how much information tinyurl.com's operators and affiliates can get from your cookies and headers...

(Everyone else here is paranoid; why aren't you?)

Re:10,000 RPM

[Anonymous Coward]

How about Redundant Array of Pretty Expensive Disks (RAPED). "We got RAPED for disks in this purchase order!". AFAIK, it's the only configuration IBM sells.

Re:10,000 RPM

[KovaaK]

Thanks. TinyURL is blocked where I work for no good reason...

Reason for restriction: Forbidden Category "Hacking/Proxy Avoidance Systems"

Re:10,000 RPM

[pjp6259]

It's not pointless if you are worried about where the tinyurl is going to redirect you.

And if you can get anything useful out of the modern internet without setting any cookies, my hat is off to you.

No price yet

[clang_jangle]

FTFA:

Samsung would not release the price of the drive because, it said, the product is targeted at the reseller market and those vendors would have to determine the mark up for the drive in their desktops and laptops.

Damn -- How can I bitch about how expensive it is when they won't even tell me!

Re:No price yet

[Capt.DrumkenBum]

If you have to ask, you can't afford it.

Re:No price yet

[bunratty]

Putting a 128 GB Samsung SSD into a Lenovo laptop (instead of the stock 80 GB 5400 RPM hard disk drive) costs $499 today. You could expect to pay at least $1000 to get a 256 GB Samsung SSD right away. Of course, the price will drop to reasonable levels in a year or two.

Re:No price yet

[wuulfgar]

Waiting for more complete data before commencing the bitching? . . . You're new here aren't you?

Re:No price yet

[PRMan]

They're not done price fixing yet...

I kid, I kid...

Re:No price yet

[Anonymous Coward]

Waiting for more complete data before commencing the bitching? . . . You're new here aren't you?

Worse than that, he RTFA.

Re:No price yet

[Anonymous Coward]

If you have to bitch, you can't afford it.

How much does it cost?

[westlake]

It makes a nice press release. But I like to see a story with a little more meat on the bones.

Re:How much does it cost?

[negRo_slim]

It makes a nice press release. But I like to see a story with a little more meat on the bones.

Apparently there is a thread going at wickedfire.com talking about how you push your adverts onto places like digg and reddit. Unfortunately I don't feel like registering with the dirt bags to find out if there is a thread on Slashdot... but with more and more posts that seem to be press releases, makes ya wonder.

Re:How much does it cost?

[jhines]

I believe the technical term is "your left testicle", but as its somewhat impolite, thats why its not quoted.

Fuzzy math

[msblack]

So it launches applications 10 times faster [sic] (should say in 1/10 the amount of time), but the article only claims speed improvements of about 3.5 to 1. People need to seriously examine how they quote or accept statistics.

Jim Elliott, vice president of memory marketing at Samsung, said the new 256GB drive can store 25 high-definition movies taking up 10GB of space each in just 21 minutes, which he said is a significant advancement over a 7200rpm hard disk drive, which takes about 70 minutes.

Re:Fuzzy math

[Walpurgiss]

the transfer speed they tout is probably the peak transfer speed. But the time it takes to launch a program also depends on factors besides the transfer speed. Like seek time for example.

Hard disks have to position the heads at the right sector before starting a read. Maybe these SSDs don't have a solid state analog to that activity and are thus faster by however long that takes.

I don't know the specifics, but I'd guess that comparing overall program access and launch time to peak transfer speed is apples and oranges.

Re:Fuzzy math

[ustolemyname]

So it launches applications 10 times faster [sic] (should say in 1/10 the amount of time), but the article only claims speed improvements of about 3.5 to 1. People need to seriously examine how they quote or accept statistics.

Jim Elliott, vice president of memory marketing at Samsung, said the new 256GB drive can store 25 high-definition movies taking up 10GB of space each in just 21 minutes, which he said is a significant advancement over a 7200rpm hard disk drive, which takes about 70 minutes.

Ah yes, but you don't have the seek times of the 7200rpm drive which are at best ~7ms. And since opening an application involves opening lots of different files (in different physical locations on the drive), this is where launching an app can be 10x faster.

So for straight writing a single, large, contiguous piece of data, it's only 3.5 times faster. For loading 200 random, tiny files, it's ten times faster.

Re:Fuzzy math

[Telvin_3d]

Both claims could be correct. It is entirely possible that by having a really low seek time and high read speed the drive could launch programs, specifically larger applications that involve many smaller files and plug-ins, an order of magnitude faster than current drives. At the same time, it could have a write speed that is only a couple times faster than normal drives. Personally, I take all of this with a grain of salt until independent benchmarks come out but the claims themselves are not impossible or contradictory.

Re:Fuzzy math

[Anonymous Coward]

"store" = write speed
"launch" = read speed

people need to read the statistics before bashing them.

Re:Fuzzy math

[MichaelSmith]

the transfer speed they tout is probably the peak transfer speed. But the time it takes to launch a program also depends on factors besides the transfer speed. Like seek time for example.

My eeePc feels much more consistent than a system with a rotating disk.

Re:Fuzzy math

[petermgreen]

It is not a contradiction for the relative performance of two devices to vary depending on the workload.

Re:Fuzzy math

[Anonymous Coward]

Yeah, except you're comparing 10 times faster (launching apps .. or reading) versus 3.5 to 1 (writing 25 HD movies). See .. reading .. writing. I still wonder about the 10 times faster, but you are mis-comparing stats yourself.

Re:Fuzzy math

[kimvette]

That the job time differs by a factor of 3.5 does not mean that data transfer speeds aren't improved tenfold. There are other factors involved, you know. It'd have been a cleaner comparison if they had transferred a single 250GB file from one HDD to another HDD, then a copy of that same file from one SSD to another SSD.

All the same, once capacities reach 750gb or better and the price point is below $200 or so, I'll be buying them. Hell, I'd probably consider buying a 256GB drive just to improve boot times. (when Linux decides it's time to fsck boot times are slow)

Question: That they could transfer 10 25 GB files to the SSD leads me to think it's 256 gigabytes rather than gibibytes? Are these SSDs rated using actual gigabytes, or gibibytes with the gigabyte label? I think SSD technology is a great breaking point where manufacturers could/should agree to abandon the misleading gibibyte ratings.

On an unrelated note: Maybe a spyware-infested Windows box will boot in under two minutes now ;)

Re:Fuzzy math

[TeacherOfHeroes]

This is somethat that a lot of people tend to overlook, either because they don't understand how a hard drive works, or because they don't stop and think about it. Loading programmes, especially ones which rely on libraries, translation files, multimedia, etc... at other locations on a disk would greatly slow down a HDD in comparison to an SSD.

Contrasted with SSDs, which are pretty much random access devices, in order to read each of those files from an HDD, there are basically 3 time factors to consider.

1. Seek time. The time it takes to move a reader head to a specific track (ring of data on a platter). Assuming that there is only this read taking place, you can pretty much assume that the reader head moves from its current location to the correct spot on the disk right away. Things are not always this pretty, though.

2. Rotation time. On average, you will have to wait half a rotation for the correct spot on the disk to spin around to the reader heads. There may be algorithms designed to mitigate this by reading even as it waits. In case the read is large enough to span a significant portion of the track, it can append that buffered data later, but I don't know if this is done or not.

3. Read time. This is the amount of time required to read the data off of a single track, and can take up to 1 rotation of the platter to complete.

So while the GP has a point in that people need to be careful about what kinds of statistics they believe, he/she glosses over the fact that reading a single piece of data with an HDD is hardly a random access, constant time operation (or linear time for n pieces of data).

Re:Fuzzy math

[westlake]

The new 256GB drive can store 25 high-definition movies taking up 10GB of space each in just 21 minutes

That will be nice to know the next time I need to manually rip 25 Blu-Ray discs to a laptop in a single half-hour session.

Rather than set up an automated download from a home media server.

Re:Fuzzy math

[oxnard]

[So it launches applications 10 times faster [sic] (should say in 1/10 the amount of time), but the article only claims speed improvements of about 3.5 to 1. People need to seriously examine how they quote or accept statistics.]

On a read there is still I/O setup time and data transfer time to be added in. The SSD reduces or eliminates seek time, ring time, rotational latency and read time.

Re:Fuzzy math

[TeacherOfHeroes]

I think you may have the Giga/Gibi lables confused. Depending on your definition of Gigabyte, Gigabytes are either the same size or smaller than Gibibytes

A Gibibyte is 2^30 (1,073,741,824) bytes. This is the only definition that exists for this term.
A Gigabyte traditionally also means 2^30 bytes, however 'Giga' is actually a base-10 prefix for 1000^3 (as used in the metric system), and therefore has a second menaing of 1,000,000,000 bytes (slightly less than 2^30 bytes).

Hard drive makers were quick to jump on the fact that they could advertise their drives as larger than they were by exploiting this double definition. If anything, what we should want is for hard drives to actually by labeled in Gibibytes to remove the manufacturer's ability to short-change customers.

Since gigabytes are the same or smaller than gibibytes, claiming that it can hold 10 x 25 Gigabyte movies gives us no guarantee that it will be properly, non-deceptively labeled.

Re:Fuzzy math

[Trogre]

I'd dispute that.

To use a car analogy, if you walk from house A to restaurant B at 6km/h and your neighbour takes the same route by car at an average of 60km/h they get there in 1/10 the time you do. They're travelling 10 times faster than you.

I don't believe there's any difference mathematically between that and waiting for an application to load. Except your neighbour perhaps should have offered you a lift.

Re:Fuzzy math

[marxmarv]

If they're smart they'll use that extra 2.4% as spares to cope better with those programs that use on-disk caches.

Re:Fuzzy math

[los furtive]

Seek time is next to zero for SSD drives (less than 1ms). There's no seeking because it's all in the register already.

Re:Fuzzy math

[Paradise Pete]

So for straight writing a single, large, contiguous piece of data, it's only 3.5 times faster. For loading 200 random, tiny files, it's ten times faster.

As fast, not faster. If something normally takes 24 seconds and you can do it in 8, then you're three times as fast. If you can do it in 12 you're twice as fast. If you take 24 then you're as fast. Certainly not 1 times faster, which would be retarded.

Re:Fuzzy math

[cowwoc2001]

For loading 200 random, tiny files, it's ten times faster.

Finally! That pesky Windows Indexing service will stop crippling my system!

Re:Fuzzy math

[maxume]

For workstation type use, as long as the wear-leveling isn't brain dead, it doesn't matter. If you assume that a disk only has 10 gigabytes of free space, 100,000 writes (which may be low by now, I don't keep track) will be able to cope with the better part of a petabyte of writes before issues come up.

Re:Fuzzy math

[Anonymous Coward]

So it launches applications 10 times faster [sic] (should say in 1/10 the amount of time),

"10 times faster" refers to the speed of launching apps (e.g. apps per hour). Saying that the speed X times faster is a completely valid way to express it.

keep it up

[gEvil (beta)]

The spinning disc is slowly beginning to wind its way down.

And of course no price

[Kjella]

IT had better not be same price/GB as Intel, at 1800$ I'd say it's a speciality interest drive. If they can match the 7-800$ for the 250GB OCZ SSD but without the horrible Jmicron controller, this could be a winner.

How many hours will it last if used as swap?

[Anonymous Coward]

Wonder how many hours this drive would last if used for swap or a database container until the flash cells wear out and start returning errors.

Re:How many hours will it last if used as swap?

[TheMeuge]

You could always use RAM for something that needs high seek speed, frequent access, and constant I/O...

Just sit the device on a massive voltage-regenerating battery-driven power-supply.

Re:How many hours will it last if used as swap?

[Kjella]

Wonder how many hours this drive would last if used for swap or a database container until the flash cells wear out and start returning errors.

10k*256GB / 200MB/s write speed = 151 days at full write 24/7. And you'll probably get some nice warnings without data loss since the typical failure mode is that they can be read but no longer written. Of course if you're using swap even nearly that much, you're doing it wrong. I'd be very surprised if my swap use exceeded 10GB/day, in which case it'll take me some 700 years to hit the write limit. And if you're running a heavy database there are drives for you, just not this one. So who do you work for? Western Digital? I think they're the only ones that haven't realized the boat is leaving and they're lost in the mountains.

Re:How many hours will it last if used as swap?

[Anonymous Coward]

Your calculation negates the fact that flash does not like re-writing. To overwrite a page you have to write a new one elsewhere and garbage collect the old one. This is what makes flash totally unsuitable for swap, and ideal for large files.

I remember reading about some sandisk invention that optimises this by reducing the garbage collections:

http://www.theregister.co.uk/2008/11/07/sandisk_extremeffs_dram_buffer/

Do samsung have similar tech? I know samsung pay sandisk millions for patent licences, do you think this is one of them?

Re:How many hours will it last if used as swap?

[Free the Cowards]

And of course every time someone levels this criticism it completely ignores the complimentary question: how many hours will a mechanical spinning-platter HD last doing all out full-speed sustained reads and writes 24/7?

At least flash drives have a predictable failure timeline, whereas HDs simply have a vague MTBF and could easily fail much sooner (or much later!) than that.

Perfect... once they become affordable

[WiiVault]

I am thrilled, as a home user I think 250 gigs is the sweet spot for my laptop. While I certainly could fill more than that, I think that mark represents a reasonable amount of space for the average mobile user looking to ditch the problems associated with a spinning platter. I also expect the price to fall quickly, making these drive much more affordable in the near future. SSD is finally getting close to the masses.

Not Yet Available?

[bill_mcgonigle]

The ComputerWorld article says "and are available for resellers today". The Samsung press release says, "announced today that it has begun mass producing". I couldn't find them in any of the usual places.

The Samsung website is particularly un-useful and hard to navigate, though I suppose it's appropriate that they require you to use Flash for this one.

SSD = Single Sided Disc

[wooferhound]

W O W
I never thought a Single Sided Disc (SSD)
would ever be able to hold so much data . . .

Re:SSD = Single Sided Disc

[HTH NE1]

W O W
I never thought a Single Sided Disc (SSD)
would ever be able to hold so much data . . .

Just try installing one flipped. Cutting the notch will surely be a bitch.

Re:SSD = Single Sided Disc

[Phroggy]

You fail at acronyms.

Re:SSD = Single Sided Disc

[Anonymous Coward]

Stupid SlashDot!

Re:SSD = Single Sided Disc

[HTH NE1]

You fail at acronyms.

"Sizzd"?

Re:SSD = Single Sided Disc

[FlightTest]

I wonder how many of today's /.'ers remember doing this. To the best of my hazy recollection, I never had a "single sided" disk fail to format both sides.

When I first heard about it, I used a second disc to mark the location and an X-acto knife to cut the slot. I recall it being several months before tools to cut the slots started showing up in computer stores.

I also recall discussions about whether spinning the disk "backwards" would dislodge dirt trapped in the liner and cause premature failure of the disk. In hindsight it sure didn't seem to.

Re:SSD = Single Sided Disc

[bunratty]

"Same sh*t, different"?

Re:SSD = Single Sided Disc

[PRMan]

We had one brand of floppies that were completely unfinished on the back and wouldn't hold data. But most of them, including very cheap brands, worked just fine.

And you didn't use a hole punch? That's what all the cool kids were doing...

Re:SSD = Single Sided Disc

[Greyfox]

I do. I never had a single sided disk fail to format, though a couple of times I accidentally hit the media while punching the notch. I always just used a regular paper puncher to punch my notches, so it wasn't quite as precise as one of the square punching tools would have been.

That didn't last all that long though. I went from cassettes in a TI 99/4A in 1984 to 5.25" diskettes for the Apple ][ machines at school from 1986-1988. By the time I got to college the 3.5" disks were starting to come out and this trick didn't work on them.

There were always rumors that treating your disks like this would shorten the life of your diskettes or your disk drive itself, but we always chalked that up to a conspiracy by the media manufacturers. Certainly my 5.25" disks lasted well beyond the point where there were drives to read them anymore. I never heard about anyone having to replace a drive because of it either.

Anywhoo what were we talking about again? The important thing was that I was wearing an onion on my belt, which was the fashion at the time...

Re:SSD = Single Sided Disc

[raddan]

I do. I also remember telling my friends that it couldn't possibly work. I'm not sure if I was dumbly missing out on getting extra storage on the cheap or if I was being smart by realizing the value of my data (which I still have to a large extent). But I was in middle school at the time, so I'm opting for 'dumb'.

Re:SSD = Single Sided Disc

[Danny Rathjens]

I used a normal hole punch. Cheap and easy. :)

Re:SSD = Single Sided Disc

[Anonymous Coward]

W O W

Yeah, but what does World of Warcraft have anything to do with it?

Re:SSD = Single Sided Disc

[Deadplant]

oh ya, and then formatting them with filesystems slightly larger than spec. good times.

Re:SSD = Single Sided Disc

[Anonymous Coward]

I paid five dollars each for my first pack of 5 1/4s (that would be the "good-ol-days"). My hole cutting tool was a regular hole punch - available just about everywhere I happened to be.

Re:SSD = Single Sided Disc

[DimmO]

haha I remember that maybe half of mine had 2 holes punched in them because the first hole was in exactly the right spot.

Re:SSD = Single Sided Disc

[homb]

Tools? totally unnecessary. You'd just punch a hole with the standard paper hole puncher thingie. It was round, but no problem. Worst case, you punched 2 concentric holes and you were done.

Re:SSD = Single Sided Disc

[los furtive]

I didn't have one of those fancy punches, I used a leather punch instead, which made tiny belt holes. My disks look like they were nibbled by a cartoon mouse, but I'll be damned if it didn't work like a charm each and every time.

Re:SSD = Single Sided Disc

[FlightTest]

Hole punch? No, my CDO wouldn't allow that. Not that I would have ever been confused with a "cool kid" anyways.

(CDO is like OCD, only the letters are arranged alphabetically, as they belong)

Re:SSD = Single Sided Disc

[justthinkit]

By the time I got to college the 3.5" disks were starting to come out and this trick didn't work on them.

Maybe not with a hole punch, but it did work. 720 disks lacked the notch that 1440's had. One end of a pair of scissors and a few minutes pressure brought you almost three-quarters of a megabyte more diskspace! Could life get better?

I wonder how much collective time was spent gaining paltry amounts of unreliable storage. How about just the time cost to format the media before it started coming out pre-formatted?

Re:SSD = Single Sided Disc

[srvivn21]

I wonder how many of today's /.'ers remember doing this. To the best of my hazy recollection, I never had a "single sided" disk fail to format both sides.

When I first heard about it, I used a second disc to mark the location and an X-acto knife to cut the slot. I recall it being several months before tools to cut the slots started showing up in computer stores.

I also recall discussions about whether spinning the disk "backwards" would dislodge dirt trapped in the liner and cause premature failure of the disk. In hindsight it sure didn't seem to.

At least on the Apple II the hole location and shape was pretty forgiving. A hole punch in about the right area would make the disk flip-able. If you missed the first time, taking another bite out would fix it. Some cellophane tape would write protect it again.

Good times.

Re:SSD = Single Sided Disc

[oPless]

> By the time I got to college the 3.5" disks were starting to come out and this trick didn't work on them.

Are you kidding ?

Punching a hole to make 3.5" SD disks (720K in DOS) in HD disks (1440K in dos) worked just fine!

However, I can, and do remember single sided 3.5" disks (really early ones), and they were not fun at all.
They were really single sided!

small, cheap, and reliable, please!

[bcrowell]

I know the impetus is to produce big and fast SS drives, but I'm more interested in cheap and fast ones. My desktop machine has 11 Gb of system and apps and <1 Gb of user files. I would be perfectly happy with a 16 Gb SSD that had great performance, was cheap, and was reliable. Reliability is a big issue. Although theoretically a device with no moving parts should always be more reliable than one with moving parts, in reality SSD technology isn't as mature as HD technology, so the failure rate may actually be higher, and there may be no way to recover from a failure.

Re:small, cheap, and reliable, please!

[david.given]

You can get 16GB SDHC cards for about 30 USD. Those are class 6, which means you get anything from about 8MB/s to 20MB/s depending on the brand. Of course, if you want more speed, you can always use RAID0.

In fact, given how cheap they are, a RAID5 system would probably make sense. You get a speed boost, and the ability to hot swap a single card if it goes bad. ZFS would also work really well, but I don't know if you'd get a speed boost that way. Also, all these approaches would allow you to very easily extend your system by buying another card (and reader) and adding it to the pool. (You may want to check up on whether you can remove it again later, though.)

Hmm. Thanks for prompting me to go and look at this stuff. I might actually do this for my next lightweight server.

Re:small, cheap, and reliable, please!

[Jherek Carnelian]

so the failure rate may actually be higher, and there may be no way to recover from a failure.

Flash fails on write, not on read. If the write was successful, then baring catastrophic failure like a bullet to the unit, it will be readable indefinitely.

Rotating disk can fail on write, but the most common failure is degradation over time that is only discovered when a read fails.

Re:small, cheap, and reliable, please!

[marxmarv]

Seems that 128GB is the sweet price point (<$300). 16GB IDE units start at about $80 by the looks of it, and I wasn't shopping all that hard.

Re:small, cheap, and reliable, please!

[AbRASiON]

I completely agree with you, although ideally I would like to have my OS, my applications, my currently playing games on the SSD.
I would estimate I could do that under 60gb now and probably 256gb for the next 3 or 4 years.

I can always keep movies / data on magnetic spinnaz! (you know hard disks)
That being said, our 'wishes' are pointless, it's not here right now, however over time, they will come inevitably, I would speculate within 24 months I'll be purchasing my first SSD and within 60 months they'll be quite common in machines.

Also of interest is an article at the techreport today about Intels new "E" series SSD using SLC and not MLC ram, it's significantly faster at writes and lasts longer.
I won't buy an SSD unless it's SLC after reading that article.

Re:small, cheap, and reliable, please!

[jez9999]

I know the impetus is to produce big and fast SS drives, but I'm more interested in cheap and fast ones. My desktop machine has 11 Gb of system and apps and

I need *AT LEAST* 10 pebibytes to store my porn collection, you insensitive clod!

Re:small, cheap, and reliable, please!

[Anonymous Coward]

I tried using a 16 GB SDHC card (class 6) as the primary disk for my laptop and it was _slow_. It works fine when copying files from a HDD but when you run your system on it, it gets a lot of CPU overhead + I think it's pretty slow at random accesses. For example, Firefox (brand new session) would start in >10 seconds (on a fresh Gentoo install on a computer with a dual core 2 GHz CPU and 2GB of RAM).
Also, I couldn't get suspend to work because the disk was unmounted on suspend.

On the other hand, I could have messed up myself. I used ext3 as the file system. Maybe a better one (ZFS as you said) would be faster?

Re:small, cheap, and reliable, please!

[XMyth]

Yea, it will incur a lot of CPU overhead. I'm pretty sure that you can't do DMA over USB.

Powers of Two

[HTH NE1]

All these flash drives and solid state drives keep advertising capacities in powers of two: 64 GB, 128 GB, 256 GB. So why do they still say a 256 GB SSD is 256 million bytes?

Re:Powers of Two

[darksaber]

If a 256GB SSD drive is only 256 million bytes, I'm a lot more concerned about the missing factor of ~1000 than the binary/decimal notation.

Re:Powers of Two

[Wesley Felter]

It has 256GiB of raw capacity but some capacity is used for overhead so only 256GB is left.

Re:Powers of Two

[Kjella]

Computer math doesn't work like regular math, like for example SATA2 which is 3Gbps. Now if I showed you a cargo ship with a capacity of 3000 tons, you'd think you could actually load 3000 tons right? And not that 600 tons of the cargo hold would have to be fixed support beams. But with computers it's somehow okay that 600Mbps is just parity bits and that you can't actually transfer more than 2400Mbps of data. And computers have been fucked with 1000/1024 at least as far back as the 1.44MB = 1.44*1000*1024 floppy which can't be right in any system and probably longer. Ignore it, honestly whoever started this has wasted more time for computer users than whoever dropped the century digits leading to the y2k problem.

Re:Powers of Two

[felipekk]

The missing space could be caused by several different things depending on your take:

. NSA backdoor
. DRM from RIAA/MPAA
. Hardware level secret pr0n stash
. Rootkit

etc

Re:Powers of Two

[HTH NE1]

I would say I must be feeling British today, except that wouldn't have excused the mistake either.

Re:Powers of Two

[HTH NE1]

If the file system's overhead consumed 17.58 GiB of my 256 GiB drive, I'd look to use a different file system. There's no cause for that much to be wasted, regardless of what your block size is.

Re:Powers of Two

[Wesley Felter]

The overhead really is necessary for bad block management and garbage collection. Some "enterprise" SSDs have 20% or even 50% overhead, so you're actually getting off easy.

Re:Powers of Two

[phozz bare]

I believe the extra space is reserved for use by the wear-leveling mechanism, not the file system.

Re:Powers of Two

[HTH NE1]

The missing space could be caused by several different things depending on your take:

. NSA backdoor

I don't think you have to worry about that. With the wear-leveling alone, you already cannot be sure whether or not something was truly erased: your rewrites may not have gone to the same parts of memory where the file was stored in the first place.

Re:Powers of Two

[iggya]

Computer math doesn't work like regular math, like for example SATA2 which is 3Gbps. Now if I showed you a cargo ship with a capacity of 3000 tons, you'd think you could actually load 3000 tons right?

No, you wouldn't necessarily expect to be able to load 3000 tons. Firstly, what type of tonnage are you talking about? In shipping, there are several different types of tonnage, or in other words, different values for the same thing, with at best slightly different names. For example, Gross Register Tonnage, Net Tonnage, Gross Tonnage, Thames tonnage, Panama Canal tonnage, Net Register Tonnage, and who knows what else.

Secondly, suppose a ship has a "capacity of 3000 tons". Could you fit more pillows or gold bars into the ship? Which one will fill the hold first? Can you fit 3000 tons of pillows into a ship with a capacity of 3000 tons? Can you fit 3000 tons of helium in? 3000 tons of depleted uranium? What if the ship is to be sailed from a salt water port into a freshwater lake? Does that affect anything?

Why would you pick tonnage of shipping as an example of regular math? Shipping measurements are all over the place. For example, how long is a ship? Well it depends on the shape of the ship, and where you measure it. Length at the waterline or length overall? With the ship loaded or empty? Heeling or sitting level? Salt water or fresh?

Just about everything to do with computers is simpler and more regular than just about anything related to boating.

Re:Powers of Two

[Free the Cowards]

My favorite example is Gigabit Ethernet, which tops out at 400Mbps for uni-directional communication even before you take into account Ethernet framing overhead, much less overhead for whatever network protocol you're using. What a scam!

Re:Powers of Two

[HTH NE1]

You don't need to reserve space for wear-leveling. The bigger you are, the more space you have to spread the wear, the longer you last. If you partition off some space to not be worn, you reduce the lifespan of the space that is being worn. And, by the time you'd need it, more space is going to fail than what you held in reserve anyway, and probably faster than you can recover it.

Why isn't wasting RAM installed in your computer necessary to make it addressable?

Re:Powers of Two

[Anonymous Coward]

That's only true if blocks fail over time due to natural write cycles. But it's also possible that they use the reserve to manage an acceptable defect rate too. So rather than spending a lot of money making sure that 100% of the storage works perfectly when the drive leaves the factory, they can spend a whole lot less and know that, for the most part, ~98% of of the storage works and the other 2% will be quickly identified by the controller and replaced with the storage that's been held in reserve.

Re:Powers of Two

[Taibhsear]

In shipping, there are several different types of tonnage, or in other words, different values for the same thing, with at best slightly different names. For example, Gross Register Tonnage, Net Tonnage, Gross Tonnage, Thames tonnage, Panama Canal tonnage, Net Register Tonnage, and who knows what else.

And let's not forget our british friend, the metric fuckton.

Re:Powers of Two

[Wesley Felter]

You don't need to reserve space for wear-leveling.

Yes you do, because flash can't be rewritten in place. You have to write into an empty block and then erase the original block. If all the space was used, you'd have no empty blocks and thus couldn't write anything. More reserved space allows better wear leveling because you can choose to write to the empty block with the least wear.

Why isn't wasting RAM installed in your computer necessary to make it addressable?

Because DRAM can be rewritten in place.

Re:Powers of Two

[Wesley Felter]

You might want to invest in better equipment, because other people are getting 9.9+9.9 Gbps out of their Gigabit Ethernet.

Re:Powers of Two

[kayditty]

I call shenanigans. do you have a source for that? gigabit ethernet is 802.3ab, or gigabit over twisted pair. I'd really like to see a system that can push that much over 802.3ab. this is copper, not fiber. but maybe I'm out of the loop?

Re:Powers of Two

[kayditty]

that should be: what this guy is talking about is most likely 802.3ab, but maybe that was your point?

Re:Powers of Two

[Wesley Felter]

Sorry, that was a typo. I meant you can get .99+.99 Gbps out of Gigabit Ethernet.

Re:Powers of Two

[Anonymous Coward]

There's a whole area that needs total revision and seems to have been missed by the think tanks.
Port & Starboard? Aft? Stern?
The nomenclature is confusing and reminds one of a 'boys, club' in approach.
Really, the whole shipping industry needs revision.

Re:Powers of Two

[Phurge]

Why would you pick tonnage of shipping as an example of regular math?

Quite right, you're on slashdot, please use either cars or libraries of congress.

Re:Powers of Two

[dkf]

Just about everything to do with computers is simpler and more regular than just about anything related to boating.

You ought to try running a production HPC service before saying that...

Re:Powers of Two

[Anonymous Coward]

You are just intimidated because you don't know the proper nomenclature. It's not a boys club thing to feed outsiders' inferiority complex, it is about precision in communication. Let's say you're backing a ship up. Which way do you turn if you turn right? Mess that up and you ram several thousand tons of steel into the quay.

Re:Powers of Two

[kayditty]

it still doesn't make any sense unless you're talking about fiber. presumably you are. going by the gist of the original post, I'd wager we're talking about home networks, where you see about 400Mbps max (typical). you may be talking at wire speed with overhead, though; I haven't used or worked with gigabit over copper very much yet.

Is it really fast?

[edxwelch]

SSDs make use of bulk read/writing to give an average i/o speed that's quite high, however reading and writing hundreds of small files that aren't located sequentially throws a spanner in the works. I'd like to see some real world benchmarks

Random access

[SlashDev]

Does a seek-time from one RAM location to the end of the 256GB (or larger drive) as fast as the avg seek time of a 15K drive? And please stop calling them disks! Disks are cicular objects.

Re:Random access

[the_other_chewey]

There is no seek time when accessing a flash drive.

Re:Random access

[FrangoAssado]

And please stop calling them disks! Disks are cicular objects.

Good luck with that. People still "dial" phones, even though phones with dials haven't existed for decades.

Re:Random access

[afidel]

Actually there is a row access time which can be quite high, as high as .5ms for MLC which compares with 2.5ms for 15k rpm drives. Add to that the relativly low IOPS for MLC (less than 100 according to this review using the database server IOMeter profile which is 70/30 read write if I remember correctly) and for a server load they lose bigtime to drives considering they get worse performance, have significantly lower MTBF, and have way less GB/$. SLC is a bit harder to quantify as the best units have MUCH higher IOPS per unit than disk so you have to figure out how much capacity you need and how reliable you want it to be as well as how much power budget you have.

Re:Random access

[Anonymous Coward]

Good luck with that. People still "dial" phones, even though phones with dials haven't existed for decades.

You've never been to Eastern Europe, have you?

Re:Random access

[Trogre]

No, discs are circular objects. A disk can be any shape you like.

Re:Random access

[SlashDev]

really? You actually believe that flash memory is instant regardless of which part of memory you are on?

Re:Random access

[Fatalis]

Hello there, reporting in from Eastern Europe. Rotary-dial phones have been uncommon here for a long time too. My suggestion would be to take what you see in movies less seriously.

Re:Random access

[bestalexguy]

Any more luck with STDs, maybe, being STorage Devices? Nice gift for a geeky Christmas Eve night ...

Re:Random access

[Anonymous Coward]

phones with dials haven't existed for decades.

Please don't tell that to my landline phone at home, it's been manufactured in 2001, but with the same parts as when they started manufacturing them in 1949.

Part Number please?

[latsamajunk]

Anyone got a part number on this?

Yep

[Sycraft-fu]

Disk I/O is the one area I still have an easy time slamming modern computers on. Most others, it isn't too expensive for me to simply get enough power that handles what I want in realtime without slowdown. Multiple VM, no problem quad cores are cheap. Big audio projects? Hell I can get 4GB of RAM for less than a month's Internet access... However when those projects start hitting the disk, I start having problems, even with a RAID array. The sequential stuff isn't it, it's the random access that kills it.

Audio only takes 172Kbytes per second per track (for 32-bit floating point). So you figure that doing something with, say, 64 tracks isn't a big deal right? Only about 11Mbytes/sec, way under what a single disk can take. However you can find that it'll choke. Reason being is that the audio isn't all nice and sequential. It's written to disk as 32 separate stereo audio files. Also you maybe have some of them reading, some of them writing and so on. The disk gets overloaded trying to seek to the information in time.

VMs are the same thing. Two VMs running computations at the same time on a system works at full speed. They each use a core of the CPU, there's no problem. The do contend for memory bandwidth, but that is plenty high enough. Likewise one VM doing disk access happens at near native speeds. There's not a lot of overhead to read and write to the disk. However get two VMs doing disk access, man things grind to a halt. Your drive is dancing all over trying to service the simultaneous requests from different areas so throughput grinds to a halt.

An SSD would just be amazing for apps like this. Not because it has so much more bandwidth, but because it's bandwidth stays much higher under intense random access. Where a harddrive might obtain 50MB/sec in sequential read, the same drive might struggle to pull even 5MB/sec in random reads. For the SSD it might be more along the lines of 200MB/sec for sequential and 180MB/sec for random. Even though it isn't full speed, it's close enough as no odds. With that, the VM and audio work would have no throughput problems.

Re:Yep

[Gazzonyx]

I've found that for VMs it's best to short stroke the drive. Partition it so that your VMs are in the middle of the drive, all together, in a somewhat narrow section of the disk. That way, even while doing high IO/sec, you're at most half a disk seek from anything else on the disk. Also, always pre-allocate your VM disks; the performance difference is huge. If you're running a *nix distro, it pays to put your swap on one side of the VM partition and /var on the other, this way you shouldn't have to stray out of the center of the disk too much. The anticipatory IO scheduler in Linux helps a good deal here, you'll get much more throughput for a sub millisecond latency cost. With NCQ/NTQ, and only using 1/3 of the disk surface, you can feel the difference... especially as compared to a single partition that's running low on space and doing a full stroke to get from the VM to your OS.

Re:Yep

[Anonymous Coward]

Early solid-state drives benchmarked at over 2000 IO operations per second.

In contrast, most SATA drives peak at around 70.

Although, of course, the size of the blocks that you're reading still matters due to cache coherency. It's still faster to read large chunks sequentially, than thousands of scattered bytes from completely random places.

Re:Yep

[shutdown -p now]

You make the mistake of assuming that the layout you see in your partition editor actually corresponds to the physical layout of the drive (i.e. something "in the middle" is actually on the tracks in the middle, etc). This is simply not the case for modern drives in general (it may be true in some specific circumstances, but as a general advice, it's worthless).

Excellent!

[zmollusc]

Perfect for systems that need to be written to once, then read lots, available with minimal delay (no spin-up) and maximum reliability. ie pr0nz server. Immense sales for this market sector alone should bring prices down.

Because that's what GB means

[Sycraft-fu]

Giga is an SI prefix. It is defined as 10^9 and abbreviated as a capital G. So to say you have 200G of something implies you have 200,000,000,000 of them.

Computers do it wrong. When computers say Giga they mean 2^30, not 10^9. That's wrong, for that you use IEC prefix of gibi, abbreviated as Gi.

The reason is that back in the day, computers had little memory. Thousands of bytes was all. So when talking about thousands of bytes, programmers started calling them "kilobytes". After all, it is close. 2^10 is close to 10^3, only 2.4% error. Well memory kept growing, and the incorrect prefix usage kept going on and they kept using bigger ones.

However this has two problems:

1) The error grows. At the giga level it is about 7% off. The large you are talking about, the more the difference between the base 10 prefix and it's "closest" base-2 amount.

2) You get confusion between levels. For example suppose your computer shows you something in megabytes. It says you have a file that is 2000 megabytes. Well that's 2 gigabytes right? Wrong, 2 gigabytes is 2048 megabytes. So it is rather unintuitive to humans. We work in base 10, the numbers displayed are base 10, but the prefixes are used wrong.

Really, the harddrive makers are right. Computers should display amounts according to the base 10 prefixes. Computers have no problems with base conversions, they should be doing that for people.

Re:Because that's what GB means

[TClevenger]

Sugar-coat it however you want, but hard disk manufacturers did it for selfish marketing reasons only. An ST-225 20 megabyte drive is about 21,000,000 bytes. A 360k floppy disk is 362,496 bytes formatted. 256MB of RAM is 262,144 kbytes.

It was only when somebody couldn't quite make a 1GB hard disk that 1,000,000,000 bytes became "good enough."

Re:Because that's what GB means

[Zerth]

That's it, turn in your geek card and be outta town by sundown.

Re:Because that's what GB means

[Burning1]

So, is a billion 10 to the 9th power (AKA a short billion, or modern billion?) or is it 10 to the 12th power (AKA: a million millions, long billion, or old English billion?)

A megabyte was defined as 2 to the 30th power because computer addresses are based on the number two. That's why addresses are typically written in hexadecimal, or octal. Measuring IT numbers this way is still extremely useful.

By attempting to re-define 'Megabyte' in a way that is more convenient to yourself, you only serve to confuse matters for everyone else. Yes, it sometimes takes a moment to explain the issue to the un-initiated, but they can typically get it. Re-defining the meaning confuses everyone unless everyone simultaneously agrees on the new meaning.

It's worth noteing, that computational sizes are very clearly separated from SI units. If you say Kilobyte, it's very clear that you're not speaking about the same unit of measurement as a Kilogram, or any other SI unit.

Re:Because that's what GB means

[Just Some Guy]

Really, the harddrive makers are right.

I'll stipulate hard drive sizes for the sake of argument, but how does that relate to RAM drives? Why do SSDs and other RAM-based devices always seem to come in powers-of-two GB sizes even though they don't really mean it? I could understand, say, a 60GB SSD being 60*10^9 bytes instead of 60*2^30. I can't understand any reason for a 64GB RAM device being anything other than 64*2^30 except deceptiveness. That's what the grandparent meant.

Re:Because that's what GB means

[bh_doc]

Computers do it wrong.

Computers aren't even consistent about it. RAM capacity? 2^10s. HD capacity? 10^3s. Network bandwidth? 10^3s. ... 1.44 "MB" floppy? Lord only knows...

Re:Because that's what GB means

[Alioth]

Speak for yourself. I work in base 16, you insensitive clod.

Re:Because that's what GB means

[ivan256]

That's revisionist history.

The IEC prefixes were created in 1999 to deal with the discrepancy introduced by hard drive manufacturers. (kilo|giga|mega)bytes had standard definitions in powers of two decades before the hard drive manufacturers started ignoring them, and yet an additional decade before the IEC prefixes were created. Worse, hard drive manufacturers used the correct definitions right up until 1995, when it was switched so they could advertise their drives as having a 1GB capacity.

Just because the prefixes are similar (identical) to metric prefixes is merely confusing for non-computer users. It doesn't mean that they're wrong.

Re:Because that's what GB means

[Sycraft-fu]

It is defined as a short billion, as in 10^9. Seriously, look it up, it isn't hard. The SI prefixes are extremely well defined, and are used for virtually all units of measurement in almost every country except for the United States.

Re:Because that's what GB means

[shutdown -p now]

Giga is an SI prefix.

That is correct. As such, it only has a prescribed meaning (that of 10^9) when applied to SI units. "Byte" is not a SI unit, therefore, SI has nothing to say on what a "kilobyte" is.

Of course, this all really matters only if you're a pedant. But if you aren't, then why are you challenging the popular and well-established convention?

WRONG

[Anonymous Coward]

You would be incorrect to think SSD's use much less power than a hard-drive. This is really visible in the netbook market where HD equipped machines have a longer battery life than the SSD versions.

Now SSD's may use less power than a 3.5" HD, but it's not quite the savings you would think and the larger the SSD, the more power it uses. I imagine this 256GB SSD draws considerable power.

Then top it off with the short lifespan of flash memory and you're still stuck with a losing solution. We tested SSD's for use in development workstations and database servers and in both cases the lifespan was on the order of a few months before the memory starting crapping out.

I hate hard-drives too but the SSD technology isn't quite there yet.

Re: SSDs vs HDDs

[Abreu]

Well, this year I bought an Acer AspireOne netbook, but I decided for the model with a HDD instead of the SSD version...

However, if SSDs continue improving and becoming cheaper, maybe my next mobile computer will be moving-part free!

Re: SSDs vs HDDs

[cheater512]

No fans? Do you want some roasted nuts? :P

RAM?

[advertisehere]

Wouldn't this make RAM effectively useless?

Re:RAM?

[neuromanc3r]

I haven't read tfa (of course), but the summary only talks about sequential read/write speeds and says nothing about the latency, which is the more important thing about RAM.

To illustrate the point: an oil tanker full of dvds can probably transport hundreds of gigabytes / second, but that does not mean that you can access random data quickly that way. Just that you can (on average!) transport much data per unit of time

Re:RAM?

[kwabbles]

SSD = 220MB/s transfer rate
DDR2 800 RAM = 6400MB/s transfer rate (peak theoretical)

Nope.

Re:RAM?

[EvolutionsPeak]

Not at all. Solid state drives are still orders of magnitude slower than RAM (i.e. main memory).

Sequential vs Random I/O

[Jjoohhnn]

Don't SS Drives have a negligible performance gain when the I/O is sequential? [or do I have that backwards, I always mess up the most the smallest details] This makes me think a SS Drive would be good for something like a windoze pagefile... but bad for something like reading a really large table that isn't indexed.

Re: SSDs vs HDDs

[Chosen Reject]

The Dell Inspiron Mini 9 has no fans. It's never felt too hot to me.

quickness != rapidness != speediness

[this great guy]

It's because you are confusing quickness with rapidness. See this drive has a quickness about 3.5x better than regular drives, but a 10x higher rapidness ! They also improved the speediness by about 5x.

Hopes that clears it up for you.

Data recovery?

[Anonymous Coward]

Can anybody tell me what data recovery is like on SSDs? Is it advanced as standard harddrives where they can pretty much get at the data even after it was overwritten?

Watch out for sustained read...

[Anonymous Coward]

I recently did some research before buying an SSD and the cheaper SSDs show some odd behaviour, namely, their reads are very sawtoothed (ie not sustained high thruput). Admittedly, the greatest speedup is in random access (no rotation/seek time), but the xfer is important too. I ended up going with a smaller, but very fast 16G mtron 7500 pro, as it's more than enough for system, swap and frequently used (db's etc). Bigger would have been nice (for warcraft and vm's), but it's not great loss. A good ssd for a system drive is a thing of beauty, and can be retained as other hardware is upgraded, of course.

Re:Watch out for sustained read...

[benow]

hmm, that was by me (grumble).

Do they have random write speeds up?

[Anonymous Coward]

I won't consider buying one otherwise.

Flash Translation Layer

[CustomDesigned]

SSDs do not allow you to directly read/write/erase flash memory. The firmware includes a flash translation layer that lets the host read/write 512 byte sectors just like any other drive. Sectors do *not* have a fixed location on the disk. Writing a sector simply appends it to the current erase block, and updates the translation table (also an append). When it runs out of blank blocks, it picks one to erase based on its wear leveling algorithm and garbage collection, and copies any live sectors to a fresh erase block. Just like a HDD, there are plenty of spare erase blocks, which are needed for the copying garbage collection and for when erase blocks go bad.

While the basic function of FTL is open, the wear leveling and garbage collection algorithms are fiercely proprietary. (The best ones actually count how many times a block has been erased and keep the counts even - and do this at high data rates.) This is OK for now because there is also fierce competition, and the code runs only in firmware on the device - not on the host. (Same as the controller code on a HDD.) Should the SSD market ever shake out into a monopoly, the basic FTL ideas are available.

Re:Flash Translation Layer

[Anonymous Coward]

the basic FTL ideas are available.

I knew we could soon be jumping to hyperspace, spool it up!

CDW

[Tumbleweed]

Good god, man, only suckers with corporate accounts have to shop at CDW.

Yeah, but when you buy enough from their lame selection of their overpriced stuff, they send you 'free' tins of cookies! We just got our yearly allotment of cookies at work. Yum. Totally worth spending tens of thousands of dollars more than we should so we get 'free' tins of cookies worth probably $15 each. (that's per tin, not per cookie :)

Re:CDW

[Thundersnatch]

Let me guess: you shop at Wal-Mart exclusively? As a corporate buyer, I'm willing to spend a bit mroe for reliable service. CDW does actually have an accurate compatability database and most significantly accurate and exhaustive stock levels. Which is very important to corporate buyers, especially when you're buying something like a Cisco switch with a million options or a $200K SAN. Having been burned many times by "oops, it's really out of stock" or "wrong SKU for description" or "price increased since your order" with Tiger, Insight, NewEgg, and others, we went back to sending most of our money CDW's way. The platinum discount is significant on software, and even without their corporate accounts you would pay maybe 1-2% more than the lowest price out there. More than worth it if it matters that your gear will arrive the next day or even same-day when necessary. Vendors with good, professional service save me and my staff time, and that time is worthmoney. If you spend an hour comparison shopping to save $100, you're a fool.

Re:CDW

[naetuir]

> Let me guess: you shop at Wal-Mart exclusively?

If I had mod points, I'd have modded you troll immediately for that comment.

Just because someone attempts to value what they are buying, and is willing to spend a few more moments than a standard corporate acquirement person, does not make them a "Wal-Mart"ite. I value shop, and I wont set foot in Wal-Mart. That doesn't mean I immediately go shop at Byerlys or Lunds or Kowalski's or whatever your local carpeted-chandeliered grocery store is.

Corps are notorious for spending ridiculous amounts of money where they don't need to. If you spend twice the price on everything you purchase, you may create loyalty with a particular company. And you may end up being able to work things out with them that you wouldn't have otherwise been able to if you shopped around. That doesn't equate to saving money. Because you shouldn't be fooled in to thinking that they're doing you any favors; They aren't. They're bending over backwards for you because you're paying TWICE the price on everything, which costs them almost nothing.

Buy ten units of X at $2,000 each. $20,000 in merchandise. Return $2000. Net: $18000.

Buy ten units of X at $1,000 each. $10,000 in merchandise. Unable to return. Buy another. Net: $11,000.

I wish corporate automatons would wake up and realize that just because it's "their business practice" does not make it better, economical, or worth doing.

Re:CDW

[Thundersnatch]

Except we're talking about price differences of (typically) less than 5% here. So your argument holds no water.

I was wearing...

[Anonymous Coward]

...apples as my kneepads at that time... but you don't hear me complaining about that, do you ?

work files are NOT accessed often.

[cheekyboy]

I bet if you did a file usage analysis, you would fine the OS files are used 100x more often than your .docs or other work files (avis).

Nothing wrong with 250gig in a laptop for work files, but it cannot hurt to have the primary os on a 16/32gig SSD card.

Any extra LARGE apps, can be install to the HD.
At least the OS can be pretty fast, and your apps wont slow down with OS usage.

It all comes down to cost/usage smarts. Why waste 80% of a SSD being used rarely. Pay peanuts now, upgrade to 4x or 8x in 6months for the same price, dont pay top dollar today, just to be COOL.

Re:work files are NOT accessed often.

[petermgreen]

Afaict most if not all of the SSDs that are comparable/better than HDDs in performance are as physically big as HDDs so doing that would mean doubling the space allocated to storage.

That is a pretty big penalty at least for the smaller end of laptops and would also make the laptop in question a pretty niche item.

Striping?

[Jeppe Salvesen]

I hope you're using striping (RAID5 or RAID10) and not doing pure mirroring or concatenation for your RAID arrays. Anyhow, I am really looking forward to SSDs that permit a lot of operations on the same bit. I deal with databases, and they too do a lot of random IO.

Re: SSDs vs HDDs

[Abreu]

The fan in my AspireOne usually turns on only when I have been using it for more than an hour in bed

When Im just idly browsing the net in a coffeeshop or some other place it never turns on

Should have pointed that out

[Gazzonyx]

Yeah... I was taking that for granted. I've got 4 500GB WD-AAKS', identically split to 9 partitions. The 'outside' (beginning of disk) two are RAID 10, the next two are RAID 1, and the remaining five are RAID 5. The very 'inside' (end of disk) is filled with dd'ed .img that I mount on a loopback to benefit from RAM page caching in the kernel. The raw images are great for virtualBox, since it's over iSCSI I can use the disks with a either Windows or Linux. The RAID 1's hold ISOs for distros, and the RAID 5's are general purpose. The iSCSI server itself is just a high end gaming board with 3 gigs of RAM and a bunch of NICs running BlueWhite64... it's somewhat hacked together, but it works (I'm a student so I had to eat a lot of off-brand Ramen to afford the rig).

At any rate, it can easily handle 5 or 6 VMs beating it down along with serving up regular files to my desktops. But I'm in the opposite side of the spectrum from you; I'm tuned for raw throughput and low IOPS... this setup would fall flat on its face with a database load. Tuning for throughput is much easier than tuning for IOPS. I'm sure you won't be missing spinning disks when you can delegate them to mass storage for backups and run on SSD at a 100th the latency.

SSD tips/tricks/techniques, WITH proofs... apk

[Anonymous Coward]

"An SSD would just be amazing for apps like this" - by Sycraft-fu (314770) on Monday November 24, @06:47PM (#25879275)

AND, they're AMAZING for:

1.) Database Servers

2.) WebServers

3.) FileServers

ALL - Per this review/test (for your reference):

----

Gigabyte's i-RAM storage device - RAM disk without the fuss:

http://techreport.com/articles.x/9312/7

----

That's PRACTICALLY applicable uses for them, on an INDUSTRIAL scale, no less... HUGE orders of magnitude of diff.!

In fact? This was their conclusion, verbatim, from that test:

"Wow. Seriously. The i-RAM is in another league in IOMeter, offering transaction rates that are an order of magnitude higher than its closest competition. It doesn't take long for the i-RAM to get revved up, either. The card hits its peak transaction rate with just two simultaneous I/O requests."

& they're right... I used the SAME techniques (albeit via a software based mirroring back to HDD ramdrive by EEC Systems/SuperSpeed.com) back in 1996 to good review in Windows NT magazine (now Windows IT Pro), & later, @ MS-Tech-Ed 2000-2002 placing as a FINALIST there, 2 yrs. straight, in the HARDEST CATEGORY:

SQLServer Performance Enhancement

By using SSD's &/or Ramdisks in Software for perf. gains & it worked...

APK

P.S.=> I use one, @ home no less (older, slightly SLOWER model, same idea though (PC-133 SDRAM + PCI 2.2 bus @ 133mb/sec., in the CENATEK "RocketDrive") vs. (DDR-RAM + SATA I bus @ 150mb/sec. in the GIGABYTE IRAM - both of which maintain nearly even/level read-write speeds, due to NOT being FLASH RAM based mind you), albeit for things like:

A.) Pagefile.sys placement (partition #1 here, @ 1gb)

B.) WebBrowser caches (partition #2, 1gb)

C.) %Temp% ops (partition #2, 1gb)

D.) Logging from the OS (like eventlogs), & apps too

The ideas I note have been "modded up" here on this website before, here (recently too):

http://tech.slashdot.org/comments.pl?sid=1014349&cid=25591403

They're ideas "normal folks" can use, to gain extra performance @ home even, via SSD use... apk

Okay?

[stonecypher]

Drives using interleaved storage methodology have already been available as PCI-e x4 cards for some time now; the FusionIO ioDrive can saturate most PCI busses, and has been available for more than a year. It also blows the doors off of these claimed specs. If I remember correctly, the ioDrive is 64 devices in parallel.

So yeah, way to go playing halfway catchup, Samsung.

Seriously?

[Gazzonyx]

I thought that was only true for 'higher' level partitioning (LVM, etc.). I partition using (c)fdisk mapping cylinder ranges to disk slices and then put the FS right on top of the partition. Do you mean that the hardware doesn't actually correspond cylinders to 'addresses' relative to other cylinder addresses (the 1024 BIOS cylinder limit non withstanding)? I assumed that this was the case from my understanding of the Linux I/O schedulers. This would surprise me, but probably mostly because I've never really considered it before.

Then again, I'm not sure that any mobo chipset/harddrive firmware that I've ever run was open source, so I probably couldn't verify that it's doing what I would assume it's doing.

Back to 5" platters?

[ResidentSourcerer]

My wife and I have just discovered the joys of digital audio on my/her itouch/iphone. It has not taken us very long to fill a 250 GB drive. People who keep movies must have a real storage problem.

The industry moved from 5" media to 3.5" media, and now 2" hard disks are coming in for more than just laptops. This has mostly come about for the sake of faster seeks, yes?

Given the disparate needs, perhaps the ordinary user needs different kinds of storage. A desktop that is running streaming video doesn't seek much. Do you care if the heads have a 20 ms average access time?

So you guys who have more knowledge about hardware than I do:

How cost effective would it be to create a 12 platter 5" form factor drive that spun at 3600 rpm, but had the same areal bit density as current 2" drives?

***

Part 2. At one point I remember 'tiered' storage. Back when disks were very expensive compared to tape. You had a set of spinning disks, and a silo full of tapes. When a file hadn't been accessed for N days, it was migrated to tape. When it was called for again, you had a long pause while the silo found the tape and loaded the file.

Are we going to have a window where this makes good sense, substituting SSD for disk, and HD for tape?