Slashdot Mirror
Intel 34nm SSDs Lower Prices, Raise Performance
Vigile writes "When Intel's consumer line of solid state drives were first introduced late in 2008, they impressed reviewers with their performance and reliability. Intel gained a lot of community respect by addressing some performance degradation issues found at PC Perspective by quickly releasing an updated firmware that solved those problems and then some. Now Intel has its second generation of X25-M drives available, designated by a "G2" in the model name. The SSDs are technically very similar though they use 34nm flash rather than the 50nm flash used in the originals and reduced latency times. What is really going to set these new drives apart though, both from the previous Intel offerings and their competition, are the much lower prices allowed by the increased memory density. PC Perspective has posted a full review and breakdown of the new product line that should be available next week."
Does this mean that SSDs aren't laughably expensive yet?
I am beginning to think that maybe Darl McBride was attacked viciously by a penguin as a child.
Fortunately I got it for only about ~$300 so I only "lost" $100 with the new ones coming out. That having been said, I don't regret the purchase at all, it is insanely faster than any other laptop drive out there, while being completely silent and power-friendly. As for TRIM support, I've heard that Intel is not going to add it for the older drives, but I'm not sure if that is just speculation or if it's been officially confirmed by Intel (Intel not expressly say the old drives are getting TRIM support is not the same as expressly denying the support). Fortunately, the drives with the newer firmware don't seem to suffer from much performance degradation, so I'm not really obsessed with TRIM anyway.
Oh and yes, it does run Linux (Arch 64-bit to be precise) just fine.
I can't wait for next year with the ONFI 2.1 FLASH chips (the new drives are not using the new ONFI standard yet) as well as 6Gbit SATA support. At that point I'll put together a new desktop that only uses SSDs, and turn my existing desktop into a 4TB RAID 1+0 file server to handle all the big files... the perfect balance of SATA & spinning media.
AntiFA: An abbreviation for Anti First Amendment.
Getting the prices lower is definitely a move in the right direction. I'm looking forward to moving to SSD in the near future, and not having to worry about hard drive crashes anymore.
You see? You see? Your stupid minds! Stupid! Stupid!
While hard drives will continue to live on for a good while yet where $/GB considerations are paramount (especially archival type applications), the performance advantages of flash drives will soon trump the decreasing cost advantage both for workstation (x25-m) and server (x25-e) environments. The case for flash in servers is even more compelling, where we measure drives in terms of IOPS and a single Intel flash drive performs 10 or 20 times better than the best hard drives on the market for a fraction of the power consumption. Understandably, many IT managers are cautious about adopting new technologies, especially when the failure characteristics are not completely known, but I suspect the advantages are so great that minds are going to start changing, quickly.
Having gotten 2 out of 3, does Intel make a trifecta here, or is there some lurking downside (e.g. limited write cycles etc.)?
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
The last page of the review states that these should cost you roughly $3 per GB. Whether that's "laughably expensive" depends on what you want to do with the drive.
AnandTech has a nice writeup too. If the price curve drops like the first-gen X-25M we should all be happy pretty soon.
I'd thought about getting an Intel SSD previously, but couldn't quite afford it. At these lower prices, I'm gonna have to grab one at some point.
While SSD may be the new kid on the block and show signs of superiority. Hard drives retain a bit of advantage over their non-moving, solid state counter parts. Hard drives can take more write overs than SSD. Flushing the cache to the actual media is still faster on HDD than SSD. SSDs are still very susceptible to static discharge versus HDD due to more surface area having sensitive parts.
I do agree with the parent. SSD are a big thing and they have some important advantages. However, let's not go putting the cart in front of the horse and say that the era of SSD is here upon us. Cost, durability, performance, and longevity are some important areas where SSD needs improvements. In some departments of each of those categories SSD wins hands down. But SSD doesn't win enough in those areas to justify the incredibly high price of the drive. So it is a bit premature to start waving the banners right now.
..and it is fantastic. This was the largest performance increase i've seen on computers in over a decade. I was going to go with a Velociraptor because I knew how important drive access latency was but then Intel patched the fragmentation issue that was worrying me.
I got mounting rails to fit the drive into my desktop case so i'm using it as my primary desktop drive for OS, some applications (Adobe Design Premium Suite runs great on it! Photoshop CS4 loads in 3-4 seconds!), and my main games. I then have a 1.5 TB secondary drive to store my data and music collection etc. I paid around $430 for my 80GB Intel X25-M so being able to get the 160GB for that same price is a fantastic improvement. I will definitely only be going SSD in my machines from now on. Everything loads faster, I get consistently fast boot times even after months of usage.
It is amazing to see Windows XP load up and then all of the system tray apps pop up in a few seconds. You can immediately start loading things like e-mail and Firefox as soon as the desktop appears and there is no discernible lag on first load like you will get with SATA drives since they are still trying to load system tray applications.
"To strive, to seek, to find, and not to yield." - Tennyson
I'm excited about the end of the tiny-primary-memory era. One of these days, maybe the line between primary and secondary storage will shrink.
"Sorrow is better than laughter, for by sadness of face the heart is made glad." [Ecclesiastes 7:3]
How can reviewers be impressed by reliability when they've only had the units for, at most, a year? When these things hit the five-year mark running perfectly well with no data loss in the home/work environment, then I'll be interested.
Ok, they may have been stress tested in factories by the manufacturers, but reviewers don't do that sort of work.
If he's the Walrus then can I be a penguin please?
All you'd need to do to demonstrate to me the greater reliability of an SSD is drop it and a regular hard drive onto the table a couple of times while they're running and see which one keeps running. That would be enough to get me impressed by increased reliability. Regular hard drives are delicate beasts.
You may not be extremely lucky to get a regular HDD to the 5 year mark, but you are moderately lucky. Lucky enough that I would recommend regular backups rather than depend on your luck with the hard drive.
Wouldn't you?
As long as they don't wear out in months, instead of years. I'm still leery of just how quickly you can start killing one of these when it's hosting the swap file. And I have yet to hear data on just how many R/W cycles 34nm cells are good for yet.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Would you run DeFrag on an SSD like you do on an HD? After all, sequential reads are still sequential reads.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
They have wear leveling algorithms. Enterprises wouldn't be buying these if they didn't work.
If you're really that worried about it just throw a velociraptor or something in your machine and put your swap file on that and use the SSD for everything else.
You are using English. Please learn the difference between loose and lose; they're, there, and their; your and you're.
When will you actually be able to buy one?
marketing. WTF?
This has been covered many times. It's a good number. I can't recall the article, but basically if you write 20GB per day, you'll get more than 5 years out of it thanks to wear leveling and extra space (SSDs actually have more capacity than they make available to you). Now, you might scoff at that but:
1) 20GB/day is a lot for the typical user.
2) People who routinely do more than 20GB/day probably need a lot more storage than SSDs currently provide (you are talking about filling the drive in 4 days) so you probably won't be using an SSD for those purposes anyway
3) People who buy into SSDs at this point in time are typically more on the cutting edge, and are likely no have moved on before the drive wears out.
4) When the drive finally does start having problems, my understanding is that it won't just fail and you'll have lost data. The failure should happen on write, and if it fails to write that will be detectable. If it writes successfully, then it should be readable. If it does fail, I believe that part will just be marked inaccessible and the data will be written somewhere else. The drive should (again, as far as I know) provide details of the failure to SMART and other disk utilities, so the problem can be detected before it progresses to a critical stage. This is much better than magnetic media, where the typical failure is that you go to read data and it is suddenly inaccessible.
Of course, this is all just what I've read about previous generations. I have no data about the 34nm, but I have no reason to suspect it's any worse.
PS. If you want to know how much you currently write to disk and you run a linux system, check out /proc/diskstats. The 10th column should be number of sectors written. Each sector is 512 bytes, so take value*512/1024/1024/1024 and you'll get the number of GB each device has written since bootup.
For me, the biggest problem from SSD is the price (on Brazil, you pay two to three times the US price) and off course, the write cicles limit too. But, if they can be cheaper, maybe now I can consider a SSD to "system" partition.
Religion: The greatest weapon of mass destruction of all time
Intel rated the first generation X25-M's at 100GB/day for 5 years, I'd be surprised if these were significantly worse.
These SSDs contain a RAM cache that's powered by the host PC IO bus. Why don't they have a battery in the SSD? The OS thinks that everything ACKed as sent to the storage unit is written, but a power failure kills the cache before it's flushed. A little battery charged off the host PC IO bus would make these drives even more reliable than spinning discs.
--
make install -not war
understand when you put them all.
In the right place.
(Yes, I know the new parts are 34 nm)
I thought the progression of feature size went: 90 nm, 65 nm, 45 nm, 34 nm.
But the graphics processors seem to be using 55, and these SSDs are being reduced from 50.
I thought they had to pour gazillions into standardizing fab construction, steppers, and all the equipment. So is some plant manager stumbling in with a hangover one morning and accidentally setting the big dial for 50 or 55 or something? What's the deal here?
My other car is a 1984 Nark Avenger.
One assumes they are MLC which are still good for about 10,000 write cycles. SLCs for 100,000.
The controller does a very good job of cycling "sectors" used, so the whole disk gets good use, rather than the same areas being overwritten constantly. The MTBF for SSDs is much higher than for conventional drives as a result, although the figure is less relevant as it's much more down to usage than anything else.
Keep enough free space on the drive for the controller to do its cycling, don't use it for constant writes (torrents are a good example of what not to use it for - as for swap file, I don't know), and the drive will last longer than you'll use it for.
The interesting difference between SSDs and platter based drive, is a write failure is not a warning sign that your heads are about to crash and you're going to lose the whole drive, it's just a failure of that one sector. Given extreme use over an extended period, the sectors would start to fail one by one, but no data should be lost, the drive capacity would start to shrink but the rest of the drive would be fine. Head crashes will be a thing of the past, thank god.
Even when the entire device's writes are used up, one should still be able to read all the data from it! :)
That's what Anandtech found out during "desktop" testing.
(And, I assume, OS, Apps and Documents loads)
That's it. 25% faster during the, what, 1% of the time your PC spends actually loading stuff off the disk ?
The rest of the time, you get nothing.
That's not worth $200 to me.
On the Enterprise front, I wouldn't know how compelling that is (or not). But on the consumer front ...
The Cloud - because you don't care if your apps and data are up in the air.
I've had to build more than one server from consumer-class components when money was tight. Once these are down to 70 cents or so a gigabyte with 500GB+ capacities - let's say in two years, if prices keep dropping as they have been - I'll be putting them in servers at first opportunity. With their random read performance, they blow away even the best server-class rotating hard drives.
I can hardly wait. Really. Rotating media is the bane of my existence.
I bought a 4GB Gigabyte iRAM box specifically for the swap file on an SSD system.
the new product line that should be available next week.
I am fighting the urge to head down to Puget Systems in Auburn, WA and see if they really have the SSDSA2MH160G2 for sale for $490.55. My guess is it isn't quite ready to be sold yet and was merely indexed by Google.
Must. Control. Checkbook.
"Giving money and power to governments is like giving whiskey and car keys to teenage boys." - P.J. O'Rourke
Part HDD, part SSD?
During operation, the SSD data is mirrored onto the HDD in the background, or, better yet, the HDD is larger and the most frequently used data is kept on the SSD but you get the whole capacity of the HDD.
Cool, thanks for the tip!
cat /proc/diskstats | grep "[sh]d[a-z] " | awk '{print $10 "*512/1024/1024/1024"}' | bc -l
unfortunately column 10 is not the number of sectors written, it is number of milliseconds spent doing IO
http://www.mjmwired.net/kernel/Documentation/iostats.txt
Unless you system is maxed out on ram, I don't see the point. 4GB of extra ram will give you the same ability as a 4GB swap file. I've never had any problems running either windows or linux with no swap as long as you have sufficient ram (under windows, the only downside is that I think it won't be able to give you any debug info if the entire OS crashes, because the swap file is where it dumps the crash log)
HotHardware has taken a crack at these new drives also today.
of doubling production costs and increasing complexity.
Boffoonery - downloadable Comedy Benefit for Bletchley Park
Yes, I'm aware of what is in that document (that's how I figured out what the columns were to begin with). That document skips over the first 3 columns of the output for it's numbering (major device number, minor device number, and device name). It considers column 4 to be field 1. Not sure why they wrote the document that way, but PsychiKiller's command above uses awk to print out the 10th column, and that does indeed give you the number of bytes written.
"Shit, my porn collection can't be deleted! My wife's going to kill me!"
When the drive finally does start having problems, my understanding is that it won't just fail and you'll have lost data. The failure should happen on write, and if it fails to write that will be detectable. If it writes successfully, then it should be readable. If it does fail, I believe that part will just be marked inaccessible and the data will be written somewhere else. The drive should (again, as far as I know) provide details of the failure to SMART and other disk utilities, so the problem can be detected before it progresses to a critical stage.
That's what I've read too, but my experience has been different. I had two of the first affordable SSDs, made by OCZ and with the infamous JMicron controller. I was having serious issues with data corruption quite soon after OS installation and I wasn't sure if it was something with the controller and Linux. I ended up using some *nix utility designed to fill the drive with a byte combination and then read it back and see if it was correct. There was apparently a multi-megabyte section of the drive that would fail writing every other bit. (Writing 1111 would read back as 1010.) Because of the wear leveling, the location of the failure on the drive would change constantly, and the drive/OS never notified of a write failure. Silent data corruption.
The drive was eventually replaced by OCZ, and neither replacement nor the other original showed similar issues so it may have just been a fluke. But it's very possible that the possibility of write failures is very much being glazed over by manufacturers. I have also seen flash devices like USB thumb drives and flash cards fail catastrophically to read written data, so I am pretty skeptical of these claims.
Still, it's not any worse than my experience with hard disk drives.
I might be in the market for an SSD soon, so I put some note together based on my reading of the articles in the topic and elsewhere. I thought I'd share them here so I can just Google them later.
Conclusions:
Also note that Kingston sells licensed clones of the X25 disks, although currently they are actually a bit more expensive than the Intel-branded ones on newegg.
If they'd use OUM for their memory modules they'd not have to worry about r/w cycles for a drive that is used for swap.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
I've never had any problems running either windows or linux with no swap as long as you have sufficient ram (under windows, the only downside is that I think it won't be able to give you any debug info if the entire OS crashes, because the swap file is where it dumps the crash log)
Unless you have the system set to share one file between swap and hibernation, and your combined swap-and-hibernation file is smaller than RAM. I've read comments in other articles telling how someone had to close programs before the computer could hibernate properly; otherwise, it would just suspend.
"Shit, my porn collection can't be deleted! My wife's going to kill me!"
Of course it can be deleted. If the unmodifiable sector is in the actual JPEG files or the directory, the controller remaps the sector to a spare.[1] If you're worried about forensic recovery, use a file system that scrambles each sector with a unique key and then forgets those keys when the file is deleted.
[1] In my experience, the manufacturer of an SSD designates roughly 5 to 7 percent of sectors as spares. Coincidentally, this happens to equal the difference between a MB and a MiB or between a GB and a GiB, allowing decimal marketing to continue. Some sectors will end up remapped before the drive is completely written because they were defective from day of manufacture.
You made me think up a good question: exactly how hard is it to do a secure wipe on one of these things with a standard NTFS or EXT3 OS on it? Because I often get businesses to donate to me their older machine when they are gonna upgrade because I refurb them and give them to the poor. They don't worry about their data because they know that all they have to do is ask and I'll perfom a DoD-7 onsite, and I have a good enough rep that they'll just hand the boxes over knowing that my SOP is to DoD-7 without even booting up one time. After all, I want the hardware, not what is on it.
But if these things become standard for businesses and they can't be reliably wiped? Good luck with refurbing. It is bad enough that so many have bought into that wives tales about needing hardware destroyed (I say show me a single site that has recovered a DoD-7. Hell show me one that has recovered from even a single random wipe. You can't because it can't be done) but if these SSDs can't be wiped thanks to the wear leveling algorithm it is gonna be another piece of e-waste filling up our landfills when they could have been refurbed like HDDs. After all, a single mom doesn't care if a PC has a 40Gb HDD as long as her kid can do his schoolwork on it.
So has anybody tried to do recovery on one of these? Or tried a DoD-7 or other secure wipe? How does the level wearing affect the possibility of data recovery? I have seen a lot of articles on this tech but I can't remember ever reading anybody testing these for questions like this. And considering these are being marketed to business as well as gamers these questions should be answered.
ACs don't waste your time replying, your posts are never seen by me.
Wear leveling does not extend the drive life in any way. It doesn't allow more writes, and it won't extend the lifetime of the drive. It simply causes it to maintain capacity as long as possible before it fails all at once. Without wear leveling you would still get the same number of write cycles, but the drive would slowly decline in available capacity.
The DoD wipe criteria are based on magnetic drives. For SSD, I believe a delete + TRIM would be equivalent to a DoD wipe, and much faster.
Don't quote me on that though.
It's not that the data can't be completely removed, in fact once data is overwritten on an SSD as far as I know there is no way to recover what was previously there. Again, don't quote me on that, but if you are re-formating a drive, filling it up with a full pass of 0's should do the trick, because unlike magnetic media it isn't "mostly 0" or "mostly 1", it's either switched one way or the other. It's a completely different technology and the drawbacks that created the need for DoD multi-pass writes to elimate the ability to read back data should not be necessary.
Once the write gates have worn out and all you can do is read it, though, physical destruction is probably the only way to eliminate the data. Even then, it would be easier and less expensive than magnetic disc destruction (not that that's all that expensive).
BTW, regarding the effectiveness of HDD recovery from random writes and the DoD wipe spec, hard drive data extraction that made the extreme measures of the DoD wipe spec necessary is very very difficult today because the drives are laid out differently than they used to be 10 years ago. In particular, standing the magnetic bits on end made the surface area for each bit an order of magnitude smaller. This is what caused the explosion of hard drive sizes (500gb drives were seen very shortly after this happened), it also required much finer equipment to read and write to the drives in the hard drive. The techniques for recovering data on drives that had been over-written relied on the fact that the small read/write mechanism inside the drives was not very precise. It did not write the perfect magnetic equivalent of a 1 or a 0, the best it could manage was a .98 or .99. Well, upon writing a "0" over a .99 (which read as a 1 to the imprecise hard drive)it became a .02 or maybe .03. Because of this the writes and re-writes were effectively trackable three or four writes deep, and with sharp equipment and a good algorithm you could reliably trace and recover the original data. Hence the multiple passes of 0's, then 1's, then random, then more 0's, etc. all designed to fool data recovery algorithms. Fast forward to modern drives, however, and the read/write heads are writing .999's and .001's instead of .99's and .01's. It is significantly more difficult to measure and recover the data these days
None of it applies to SSDs though, they are different in every way and the DoD spec is meaningless when applied to them. Seriously though, if you want to be truly secure, a delete + TRIM should be enough, but there are also already programs that will fill the disk with useless data, after which a delete + TRIM would be as secure as you can get.
Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
That and I don't believe the data itself is lost is it? Only the capacity to erase the data and re-write it is lost? In other words, it would essentially turn that it into a read-only device in those 'bad' blocks?
Why does everyone insist on using cat where grep alone will suffice?
Everyone likes cats!
But mostly, it's just what they're used to. There's nothing wrong with it, it's roughly the same efficiency either way, so do it the way that makes sense to you, and for the problem you're solving.
"It's always fun to read bleedin' edgers rationnalize how they didn't pay over-the-top for immature first trys that soon got obsoleted." - by obarthelemy (160321) on Thursday July 23, @03:33PM (#28799041)
I've been using SSD's (& software ramdisks before these, since 1991 or thereabouts from DOS, thru Win9x, thru NT & later physical SSD's 2000/XP/Server 2003) since late 2002... & the type I use is NOT based on "FLASH" ram (w/ its limited lifespan + slower write speeds), & are mature.
First: In the CENATEK "RocketDrive" (PCI 2.2 bus @ 133mb/sec transfer rate, 2gb PC-133 SDRAM)...
Secondly/more currently/lately: In the GIGABYTE IRAM (SATA 1 bus @ 150mb/sec transfer rate, 4gb DDR-400 RAM)...
----
"I've got some oil here that will do wonder for your hair ! it is expensive, too" - by obarthelemy (160321) on Thursday July 23, @03:33PM (#28799041)
I can tell you, AND ALSO SHOW YOU, right now (& prove it) that these units work FAR faster than ordinary mechanical HDD's - Especially for seek/access (many orders of magnitude faster here by FAR) & READS (nearly 4x - 5x as fast here in fact, & in a test I had done using HDTach 3.x), which is MOST of what MOST USERS, do (reads, vs. writes) most of the time.
Results of said test, for your own reference &/or verification, vs. your statement I just quoted there above now?
See here, to verify my statements -> http://www.thenewtech.com/forums/storage-removable-media/interesting-comparison-i-did-2-diff-true-ssds-7819/
The results there should function to bear out the truth of my statements here... &, possibly, to "justify the expense" of this current crop of FLASH RAM based SSD units to others, because I was out to do so myself on the IRAM here (which the larger & more current FLASH RAM based units are more expensive than the type I use (@ least lately the current FLASH units cost more, but, these FLASH based units have a FAR LARGER storage capacity vs. the kinds I use))...
Also, that test of mine MAY possibly also function to give others some "unique ways" to utilize these units as well on their own systems, other than for a main bootup disk etc. et al! I apply mine for both read AND WRITE in nature tasks (because the type of SSD I use also does well on writes, whereas FLASH RAM based ones do not by way of comparison to the type I use).
I say this, simply because in that benchmark "drag race test" I did between the 2 units I use here, I also extoll the methods & means for which I use them for @ home for superior overall system performance
(E.G.-> PageFile.sys placement, %temp% & %tmp% placement, webbrowser cache location, spooler location, logging location for apps & the OS as much as possible to offload those tasks from my main C: drive (in essence speeding that up to by doing so) & to offset fragmentation those smallish files can create in their own selves + other files too, a "longer term" overall performance gain, in other words on that last note).
Also, many years earlier in 1996-2001, though I do not mention it in that test? How to use them for gain in "industrial environs" as well... this I outlined for BOTH CENATEK in a review I did for their RocketDrive unit (used to be on front page of their website in fact, until the server it was hosted on went down, not mine)... &, for EEC Systems/SuperSpeed.com, @ Ms TechEd 2001-2002 iirc, for my ideas for "creative uses of ramdisks", albeit in "industrial environs" also!
Which, again, took a company called EEC Systems/SuperSpeed.com, whom are a certified MS partner (for whom I did some work on paid contract for in regards to software based ramdisks & diskcaches they produce (the type that works @ diskdriver level, NOT filesystem levels where I made it up
It should be longer than that actually -- using the 10k write lifepsan and a 80gb drive, your life span would be:
(10k * (80gb/20gb)) * Q
where Q is the efficiency of wear leveling and defending against write amplification (as Intel calls it).
using Q of 0.5, we get 20000 days, or 54.8 years.
I think it will be less than that though...10k writes makes a lot of assumptions, and I think is only guaranteed for a particular period of time....the flash cells 'age', and the age interacts with write lifespan.
I can't find the website where I originally read the numbers I quoted, but that website showed the calculations similar to yours, and they came up with the 5 year, 20GB figure. However I did find the numbers I quoted straight from Intel:
http://www.intel.com/cd/channel/reseller/asmo-na/eng/products/nand/feature/index.htm
Expand the "Comparison chart" link a few paragraphs from the top. You will see an additional table. Last row in the table:
"5 years - 35TB written, up to 20GB/day for 5 years"
I'm still leery of just how quickly you can start killing one of these when it's hosting the swap file.
Then you should probably buy more RAM.
I bought a 4GB Gigabyte iRAM box specifically for the swap file on an SSD system.
That's a damn expensive way to fix problem(s) that likely didn't exist in the first place...
Interesting...I wonder what accounts for the difference in the numbers...aging, or just being conservative? Or maybe the 10k figure is not accurate...using it, an 80gb drive should be in theory able to write 800tb, not 35. Probably, all of the above.
Truecrypt has a handy feature where it can tell you how much data has been read and written to an encrypted system drive.
My system has been on for about 45 minutes, doing some email, web browsing and RDP. Total read data is ~550MB and total written is ~179MB.
Besides, by the time the SSD fails (which is more likely to be down to some kind of electrical fault or failing on-board flash/RAM) you will be able to get a new one for a fraction of the cost.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
Well, your 800tb figure would be based on a perfect distribution of writes...essentially like treating the entire 80GB as a gigantic ring buffer. Factors that could modify that figure:
1) First, your very basic figure should be slightly higher, because these 80GB drives actually contain something like 6GB additional capacity that is unavailable for use. It's there so that the wear leveling has some extra room to work with.
2) You have to write in pages at a time, but there are going to be many times when you don't need to write a full page, so that's the first thing to cut into the theoretical amount.
3) Wear leveling isn't going to be perfect. There are going to be pages that are written to "unnecessarily" as part of the erase process. With SSDs, you can write in pages but you have to erase in blocks, so to modify a used page, you copy the entire block to cache, modify the page in cache, erase the entire block, and write back the modified block. How many writes are "wasted" like this depends on both how you write data and how intelligent the wear leveling algorithms are.
"That's a damn expensive way to fix problem(s) that likely didn't exist in the first place..."
Not really. I got a good deal on the chassis, and pulled the RAM out of my box of junk.
Plus, half the reason I did it this way was just to see how well it would work.
But damn, is it fast! Paging basically incurs no performance penalty at all.
Does anyone know a store that sells this?