Slashdot Mirror
Seagate Announces First SSD, 2TB HDD
Lucas123 writes "Seagate CEO Bill Watkins said today that the company plans to put out its first solid state disk drive next year as well as a 2TB version of its Barracuda hard disk drive. Watkins also alluded to Seagate's inevitable move from spinning disk to solid state drives, but emphasized it will be years away, saying the storage market is driven by cost-per-gigabyte and though SSDs provide benefits such as power savings, they won't be in laptops in the next few years. A 128GB SSD costs $460, or $3.58 per gigabyte, compared to $60 for a 160GB hard drive, according to Krishna Chander, an analyst at iSuppli. 'It will take three to four years for SSDs to come to parity with hard drives,' on price and reliability."
Every news source has merged those two statements together, and every time, my brain gets stuck on it.
Seagate is announcing two seperate products. One is a SSD and the OTHER is a 2TB hdd.
Modding Trolls +1 inciteful since 1999
This
I can't help it. I remember buying my first two gig drive for $780, back when the dollar was worth squat. Now of course, the value of the dollar is rapidly approaching diddly-squat.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
The SSD from Seagate is a total "Me Too!" product. Seagate thinks they are in the "Mass storage" market, but they are not. They never have been. Their market is the one that includes "rotating magnetic platters". The only reason they are trying to break into this market (that they continuously decry as useless, futile, and too expensive) is because they are afraid of what "might" happen ten years down the road.
It's so nice to see a company that fought this at every step pretend to embrace it.
Modding Trolls +1 inciteful since 1999
Technically speaking, if it were always about price vs performance, we'd all be running last generation AMD's, using CD-R's and the like. In reality, you don't pay proportionally more for extra performance, you pay EXPONENTIALLY more.
/. readership, a 20% boost in performance is worth a 200% increase in price, especially considering how cheap computing equipment is these days, compared to the utility it offers.
For the average consumer, SSD's aren't yet the way to go, but for what I'd bet is a good proportion of the
which utilizes both SSD and a mechanical disk to get the best of both worlds in a way similar to processor caches L3 >> L2 >> L1. Ofcourse, current drives already use buffering but the buffer data gets lost when the drive is switched off.
They called me mad, and I called them mad, and damn them, they outvoted me. -Nathaniel Lee
I've told you a million times not to exaggerate...
Is it that fucking hard to include the cost per gigabyte of the current hard drives ($0.375/GB for the example given)? Why quote one $/GB figure if you can't be bothered to include the other?
[b.belong('us') for b in bases if b.owner() == 'you']
SSD will never reach parity with hard disks because of the economics of spinning disk storage. Yes folks, a 160GB drive costs $60.
SAMSUNG Spinpoint F1 HD103UJ 1TB 7200 RPM SATA 3.0Gb/s Hard Drive - OEM. Cache: 32MB. Form Factor: 3.5". $184.99
Seagate Barracuda 7200.11 ST31000340AS 1TB 7200 RPM SATA 3.0Gb/s Hard Drive - OEM. Cache: 32MB. Form Factor: 3.5". $209.99
Next year these will be 4TB, 8TB, 16TB? $100-$200 range. Call me on it; by December 2009 (i.e. in 2009, next year) it'll happen. Where will we see the SSD price point?
Support my political activism on Patreon.
I'd buy a $460 128 gigabyte SSD in a laptop. Not to long ago these options were about $1000. If you do this right (and often it's been done wrong) you get better performance, much longer battery life, and enhanced reliability. With the right software monitoring of repeated writes, you could also know about hard drive failures coming in advance. That's fantastic, in my book. $460 is still a tad high, but I'd bite.
Christ, as if existing standard hard drives aren't unreliable enough already.
:p) modern IDE/SATA drives fail within 3 years as is, they already have a pretty high failure rate.
It's not unrealistic to see 1 out of every 10 to 20 (well, 1 in every 3 if you use Maxtors
I thought one of the major advantages of SSDs was their reliability or is that simply not the case? are they really so unreliable currently?
One of my biggest dislikes of hard drives in general is reliability, I want to be sure my hard drive wont just not work one day leaving me without my files and backing up is one of those necessary evils as is. I build RAID in to all my machines nowadays with redundancy but I still feel like I'm getting screwed having to buy multiple disks and only get a portion of the total storage.
Would it be nice to be able to save files one day on a standard consumer system and be guaranteed they'll always be there the next without ever needing to back them up and without having to buy at least twice as many drives for redundancy.
I remember my long-former managers happily paying nearly $10k each, for the damned things...
Quo usque tandem abutere, Nimbus, patientia nostra?
Spinning hard disks will go the way of tubes in ten years, more likely faster than that. Scaling the manufacturing up will drive SSD drive costs down. There are long-life reasons why:
- SSDs aren't as vibration sensitive (both will not take a bullet, but only SSD can likely survive a normal drop of 2M on to concrete)
- SSDs don't have the temperature/altitude constraints
- SSDs don't have latency and no rise/shutdown time for green needs, in fact, they use hardly any power at all
- SSDs are generally faster, although there are algorithms needed in flash to prevent bucket overuse because reads are almost infinite, but writes are not
- SSDs take less in terms of precious metals and present fewer QA problems
- No electromechanicals to wear out.
The price point? Going down. It's an obvious solution to a long time problem. Magnetic versus flash storage will tend to favor flash, as magnetism decays sooner than flash will-- when flash is written to correctly.
---- Teach Peace. It's Cheaper Than War.
For two reasons. First and foremost, low power consumption. Secondly, we have already passed the sweet spot in the storage capacity needed for the applications most people run, particularly on laptops. Add to that the fact that current HD form factors are an extremely good fit for SSD units, and the writing is on the wall.
So what will happen is pretty obvious. Laptops are going to push SSD storage into the mainstream, giving it the critical mass needed to start the research bandwagon rolling, and 5-10 years after that happens hard drives will become the 'new' tape storage and most production systems will be using SSDs.
Even more pointedly, with power costs being the premium concern for data centers these days, and the hard drive being the only thing left in the computer that can't be engineered down to near 0 power consumption when idle (short of spinning it down, which has its own problems), my expectation is that large commercial concerns will see a huge cost benefit to using SSD storage despite the higher front-end cost of purchasing it.
-Matt
The false dichotomy is that it's an either/or thing.
There will be SSD components with high speed and low power and their price/GB will decrease very quickly. Largest capacities will always be expensive. For a long time they will cost more than magnetic media, but it probably won't always be so. Their speed and reliability will improve as vendors build out the drive intelligence that abstracts the physical media from the logical media and parallelize atomic access with internal RAID to compensate for the slowness of individual cells. These products will sell to users who are interested in their benefits and the manufacturers will make lots of money. Ultimately the speed of random reads and writes of SSD media will be limited only by the interface as solid state components are "always on" and each block of data is as accessible as any other.
Magnetic media will continue to drop in price as well. As storage increases today's hard drives will find their way to the recycling center in record time. The optimal price/performance will continue to improve as will maximum capacity. Speed will not increase as much, particularly with random reads and writes, because the data is still stored on a rotating physical object and a physical read/write head must move to the correct track and wait for the data block to fly under the head before data can be read or written. These products will continue to sell well for a long time to users interested in their benefits and the manufacturers will continue to make lots of money.
Both will be popular for a long time. There are other technologies in the works also.
Ultimately at 60MB/s it takes 1,000,000 seconds (11.5 days) to write 60tB to a (currently theoretical) rotating platter drive. At 6tB/second (interface TBA) it takes 10 seconds to write the same data to a solid state device. The ultimate winner in this one is clear, but it will be a long time.
Let me be the first to say: "that's a lot of porn."
Help stamp out iliturcy.
Hello, Can someone please explain to me how spinning drives are "more reliable" than SSDs? I mean the ONLY explanation I could give just that regular HDs have been used for so long, so we "know what to expect".... 'It will take three to four years for SSDs to come to parity with hard drives,' on price and reliability."
Hutz: All right gentleman. I will take your case. But I will require a thousand dollar retainer.
Bart: A thousand dollars. But your ad says "no money down".
Hutz: Oh, they got this all screwed up. [corrects ad with felt-marker]
Bart: So you don't work on a contingency basis?
Hutz: No, money down. Oops, I shouldn't have the Bar Association logo here either.
Since when does any 128GB SSD cost only $460?! Taking a look at newegg, the cheapest 64GB SSD is 900 DOLLARS. Newegg only sells one 128GB SSD right now, and it is three grand.
The analyst mentioned in the story apparently does not have her facts straight, or he/she was misquoted. The figure is off at least by a factor of four.
Where are they finding 128 GB SSD's for $460? The lowest price I've seen on these is around $3k ($3,049 at NewEgg, $2980 at eWiz). That puts the cost/GB at around $23.50/GB, give or take.
:P
I'd jump at the opportunity to buy one of those at that price, if only to turn around and sell it on eBay.
That green slime had it coming.
Don't worry, I'm just getting out of grad school and I still have a couple of Micropolis drives running here :)
I'd buy a 10gig solid state drive, if you're looking for the performance gains then that's big enough for either one of these: XP install, vista install (maybe), Oblivion, Crysis, etc.
I'd pay for a 30 gig SSD, I'm sure alot of other people would too.
...paying $400+ for a 128gb SSD to replace the standard sata drive in my laptop as long as the performance was truly better and the battery life was that much better.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
It's not impossible to implement that functionality with a dumb SSD and HDD. The easy part is unionfs -- done. The hard part is determining with sufficient accuracy what files are unlikely to be written again -- a first cut could just consider some directories, MIME types and/or file extensions more or less likely to be rewritten than others. The ugly part: file metadata has to be present for both file sets at all times (or at least all directories which are split across both devices), metadata might be changed frequently, the HDD must be on for as little time as possible, and writing to flash must be avoided as much as possible. The only way to satisfy all those constraints is by reading and maintaining a complete write-back cache of the HDD's inodes and dirents in RAM at mount time, flushing dirty entries whenever the HDD spins up and writing through whenever the HDD is on. At 144 bytes apiece a cache for a typical homedir/archive disk could eat up a sizable chunk of RAM. While we're dreaming, database engines could even be optimized to read only from the SSD-portion of a hybrid drive if a particular data point had not been written to in over N minutes, or since the last collation (explained later), but would write to the platters, and then during quiet cycles, it could do a collation. The collation would move data which was on the platters, but which did not have a pattern of large volumes of writes back to the SSD volume. An equal amount of battery-powered RAM as cache and journal for a traditional HDD would under most real workloads beat RAM+SSD or HDD+SSD. If you really wanted to identify (manually or otherwise) cold tables and load them into flash SSDs, the database engine will probably still load and cache them in main memory anyway (costing all of a few extra milliseconds), and any RAM not used to cache those tables can be used to speed up temporary tables or for dynamic caching. (compare Amdahl's Law) And... I'd like a pony... NOT YOURS
The usual structure of a storage hierarchy is that each level contains a fast, small subset of the next level. A consequence of this is that at the steady state the final level contains a complete copy of everything. Poor write endurance makes flash SSDs poor participators in this sort of hierarchy.
/. -- the Free Republic of technology.
Yeah, well, we know that college students are poor and thus work with old, outdated, often donated equipment.
My blog
Working at a data center myself, I'd say that heat might be more of a concern than power, but SSD trumps existing platter-based media in that regard as well. Something that doesn't move is much cooler than a few platters spinning at 15,000 RPM, after all.
For many applications, what matters is price per IO operations per second. Sometimes price per IO operations per second per U of rack space or per watt is what matters.
Flash will beat hard drives there much sooner than it will beat hard drives in simple $/GB.
Flash has fast reads and slow writes. Sun is promising flash drives with 25% of the space reserved for bad block remapping, and a huge amount of supercapacitor-backed write cache. They are promising to release this in SFF HD form factor with a SAS interface, and SO-DIMM form factor with a presumably proprietary interface.
Even if Sun breaks its promise, as is typical for them, someone else will come out with the product.
is very dependent on the application. In particular it depends on the mix of linear vs random operations and the mix of reads and writes.
For 100% read applications SSDs tend to be similar in performance to hard disks when reading linearly, and a lot faster than hard disks when reading randomly. This shows up in linear read speeds of 100 MB/sec for a typical Flash SSD which is "close" to a hard disk. For random 4K reads, Flash SSDs can stomp any hard disk. Most disks are in the 10,000 4K read IOPS range where 15K SAS drives are in the 250 range or 40x slower. So for applications that are 100% read SSDs can be as much as 40x faster, although the average is usually in the range of 15x to 20x.
When you start writing to Flash things get interesting. Flash is really designed for large, linear, aligned, writes. With most drives, you can get maximum write throughput only if you write exactly aligned with the drives internal erase blocks. Thus you can write exactly 2 megabytes on exact 2 megabyte drive boundaries and get 100% of the theoretical write throughput of the drive. Unfortunately, no application acts like this, so you are at the mercy of the file system and Flash controller to turn your smaller, probably random, and probably mis-aligned writes into what the drive can handle. The net impact of this is that good Flash SSDs have 4K random write IOPS in the 120s which is 1/2 the speed of a 15K SAS drive. I have measured Flash SSD with 4K write IOPS with values like 135, 120, 64, 43, 24, 13, 4.0, and 3.3.
This is why Flash SSD performance is so hard to judge. The random write performance can suck up the available "drive time" and dig a system deep into dirty buffer flushing. We talked with one Dell laptop user that described their system becoming "unusable" while an Outlook indexing operations was randomly updating a big file. Unusable in this case was 2+ minutes for to bring up task manager.
These random writes also have a real impact on the wear of the drive. Every time you seek a write, you basically chew up a write/erase cycle, even if the write is only 4K long. If you look at a drive that claims 50 GB/day for 10 years, this is 50 GB of linear writes on exact erase block boundaries. If you write 4K randomly, the 50 GB really means 25,000 4K writes or 100 Megabytes of random writes.
The solution to this is to not write randomly to the drive. There are file systems designed for Flash that address these issues. These are typically called "Log File Systems". Unfortunately, there is no generally available file system really designed for performance. In Linux the LogFS options are really tuned for small memory small storage systems and for hardware where the flash chips are directly accessible. They do help drive wear a lot, but they are just not tuned for Gigabytes of space or database crunching performance.
Another solution is my companies product called MFT (Managed Flash Technology) which is a software block mapping layer that runs on the host. It gives you the random write performance benefits and wear benefits of a LogFS while allowing you to use whatever file system you wish. MFT was developed on 2.6 Linux and has been ported to Windows. With MFT, the same drives that do 25 4K random write IOPS usually measure over 10,000. The linear speed of the drive is still equal to a hard disk, but the random speed is now closer to symmetric with reads and writes. Thus jobs like updating databases can literally run 20x faster than the fastest hard disks.
In the end, Flash SSDs will find specific markets initially. I can say with certainty that they won't get used for off-line backups or storing/edit large quantities of HD video. But give them databases or file systems with lots of small files, and they can really smoke a hard drive.
The most attractive features to my mind are the fast application/OS loading times, and shock survivability, at least for the portable market segment.
If I have seen further it is by stealing the Intellectual Property of giants.
Rack mountable storage? HAH...
Unless you refer to it as DASD, you're just a kid. I cut my teeth on IBM 3370. http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3370.html
The next step is to stop treating them as "disks". We tolerate the library and OS overhead of getting to a block on a disk drive because access times on disks are so long. But solid state memory devices can be accessed in microseconds. We need a different model for these devices.
Does this mean reliability will go down?
Somewhere I still have three Wren III *dual head* drives and a Micropolis...
I still use the case that those four monsters were in (for those who should get off my lawn: Each was 1 5.25" full height drive (same as 2 cd-rom drives)). Difference is rather than storing an amazing 160*3+720MB (1.2GB Baby!), I now have 4 3 drive enclosures for a total capacity of 12TB (currently at 6), and if I wanted to do away with trays I could fit 20TB in the same space (but cooling would be a bitch).
-nB
whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
The EeePC is a fairly high-power device by mobile standards. The CPU takes around 5W, while a 1GHz ARM chip takes under 250mW. The difference between an Intel CPU with a mechanical disk and an ARM CPU with a solid state disk is almost 1000%, which translates to a big difference in battery life for a portable device.
I am TheRaven on Soylent News
With hot file adaptive clustering ( http://developer.apple.com/technotes/tn/tn1150.html#HotFile )
Basically, read only often accessed files are moved to the zone on hard drive where the access to files is fastest.
It would not be hard to adapt this behaviour to move the files onto SSD portion of the disk at all.
As the island of our knowledge grows, so does the shore of our ignorance.
GAAAAAAHHHHH!!! Please, if you're not going to use any articles, leave out the fucking commas. In fact, proof-read the titles.
And, might, I add, SSD, 2TB HDDs make, my dick hard.
Please stop stalking me, bro.
How come we never hear the fact that SSD is limited to 100k write cycles. Which makes it great for data retrieval but use it for something like swap space or browser cache and its going to die quick. I like thumb drives and all but all the hype that ssd's are the magic bullet confuses me or is this really just not a problem?
Probably because the 160gb drive's $/GB figure doesn't look so great.
Heck, I've bought 500gb drives for that price. Only problem was, they weren't Seagates.
-Billco, Fnarg.com
The US Army is already buying a bunch of toughbooks, from panasonic, for various field applications. I was surprised to find out the the toughbooks use solid state drives. Take that for what it's worth. As Robin Williams, in "Good Morning Vietnam," said, "If it's being done right, it's probably not being done by the US Army."
I work as a sysadmin at an ICT department for several broadcasting organisations.
:) it is very unlikely that local storage needs will wane the coming years. I see multi-TB at home becoming the norm and unike laptop-drives, these will be full to the brim with:
:)
We're seeing what media-storage needs we have right now and our department is making predictions on future storage needs based on the plans of these broadcasters. For this, admittedly niche, application SSD's won't do for many years to come. We're talking about 200TB+ of storage right now and expanding at an alarming rate.
This becomes relevant at the consumer level as HDTV is gaining acceptance and consumers everywhere are embracing audio/video applications. Unless the network scales up, people will want to store those movies and whatnot on their home computers.
Given the current problems ISP's seem to have with coping with the amounts of data streaming television is causing, (and we see this first hand in difficulties over interconnects and peering, being able to deliver over 25% of the total available interconnect bandwidth in concurrent video-streams
- home videos (HD)
- bluray movie rips
- HD audio registrations
- HD television recordings
- etc.
Google will make a killing in offering software to do meaningfull indexing on those amounts of home storage
Karma? What's that again?
"Watch! I can drop it and it still works!
If you take out the most fragile piece in the system, then the next most fragile piece becomes the problem, like the video connection to the motherboard, the flimsy plastic case it's in, or the LCD. If you're dropping your computer, you've got more problems than just losing some data.
Besides, "while operating?" In what instances do NORMAL people need to be using a laptop while being subjected to G-forces? Riding motorcycles? Rollercoasters?
The advantages are niche, and not mainstream. People are using more space and want it cheaper. They don't want less space and more expensive as a trade off for features they don't rely upon.
Corrections: There are Linux log filesystems tuned for gigabytes of flash now. Look up UBIFS and logfs. There is also a variant, I think of YAFFS, used by some database engine.
Current Flash SSDs act like flash but present a block interface. UBIFS appears to require raw flash access thru UBI and YAFFS 1&2 also expect to drive the flash directly.
LogFS can use block devices but does not appear to be "production ready".
I am sure that people will start paying attention to the new flash drives, both from the software and the controllers point of view. MFT has the advantage of working with the hardware and software that we have now. It is also doing some very unique things in terms of free space management, migrating user blocks into usage zones, controlling wearout rates of drives, etc. These are all things that our patent writes and lawyers have been careful to state in our patent paperwork. In fact, the whole flash arena is a minefield of patent stuff, so new flash FS work could get "interesting".
You should also realize that MFT is targeted for hardware configurations that just did not exist a couple of years ago. Specifically, MFT assumes you are willing to burn quite a bit of kernel memory to make the storage run fast. This works out to about 1.2 MB/GB of usable storage. While this is 100% practical for "big servers" with performance critical database, and even quite reasonable for workstation and laptop usage, it is a show stopper for many embedded device environments.
So at this point if you want 1TB of fast Flash SSD storage with good random write performance, you can use STEC Zeus IOPS drives on a EMC array for some wild price (probably approaching $500K) or regular Flash SSDs with our software layer. 1TB using the new 1000 series drive from Mtron would be:
20 64GB SSDs $ 9,560
Chassis $ 1,000
Raid Ctrls $ 900
MFT License $ 8,250
-----
19,710
This is 18 drives raid-5 w/ 1 hot-spare. Performance is about 60,000 read IOPS and 50,000 4K write IOPS. Linear speeds on reads top out at 700 MB/sec. Basically you hit Linux raid and controller latency limits long before you saturate the drives. Usable space is 1036 GB in the configuration. The 1000 series is not quite shipping yet, so the largest array like this actually built was assembled by one of our resellers with 22 32GB Mtron 3000 series for about 500 GB of usable space. When you run benchmarks of these against "normal" drives, you get jaded very quickly.
Current "high endurance" flash memory blocks are rated for a few million erase/write cycles. Run vmstat for a typical 24-hour period in your own system, assume perfectly level write cycles (neglecting the reality that file system metadata blocks will probably be written far more often and thus wear out quicker), and solve FlashCapacity*WriteCycles/BlockWritesPerDay to find out how long that would last at most.
Flash is not battery-backed RAM, and really, battery-backed RAM is only a few times more expensive. Oh, and it writes about an order of magnitude faster.
/. -- the Free Republic of technology.
It would have to be fast for that kind of money. $20,000 for 1TB? A 1TB hard disk costs $200.
Granted, I'm sure you're right and it probably is that fast. There's certainly a niche for it.
(I have some relevant curiosity as I'm developing a database engine.)
Your license seems rather expensive, but I'm not one of your customers. Still, it's a motivator for other developers to undercut you.
From your basic description, it seems likely that UBI + BLKUBI (block driver over UBI) may give similar functionality to what you described but without the performance, because BLKUBI doesn't group 4k writes consecutively. (There are people using that combination in some embedded products in development now.)
However, UBIFS + loop mount, or LogFS + loop mount, might get close in performance. Both of those present a block device which can be used by any filesystem, but underneath translate to a logged tree-structured filesystem.
UBI is too slow at mounting huge drives, but LogFS is good for that.
Does your system offer a significant difference from UBIFS/LogFS (containing a single file) + loop mount?
I suspect BTRFS (similar to ZFS) will start getting good at this eventually. It's not aimed at flash right now, but it's a natural direction for a log-tree-structured filesystem to go in when SSDs become more popular. And it has better RAID than drive-level RAID in many ways.
Granted, I'm sure you're right and it probably is that fast. There's certainly a niche for it.
(I have some relevant curiosity as I'm developing a database engine.) 1TB of raid protected SSD for 20K was un-imaginable just a few years ago and without MLC commodity drives plus MFT it still a ways off now. Remember that 50,000 IOPS is 200 15K SAS drives. By the times you raid-10 them, put in controllers etc., and even if you "do it on the cheap" (ie from a SAN vendors this will cost a lot more), you are talking $80K and you only get 3.6TB out of this (200 raid-10 36GB drives). Lot of GBs at 100 IOPS is easy. When you need 1000 IOPS it gets a lot harder. Your license seems rather expensive, but I'm not one of your customers. Still, it's a motivator for other developers to undercut you. This is what we are selling at. Our licenses have actually come down over the last 6 months. Our policy is to follow the drive prices down. We also work with good reseller discounts. I also suspect that you don't have a lot of contact with the enterprise storage space. Things like a "volume management option" on your SAN can cost $10K. Remember that regular FC drives for EMC servers are 3x the "stock" drives cost because of the special firmware.
Other developers will probably come along, but they need to be very careful. The whole area of Flash is a minefield of patents. We have been very careful to keep the scope of what we are doing compartmentalized. We also pay our lawyers very good money. Even though some platforms like Linux have the attitude of "that looks good, I can write it to", the business world does not work that way. From your basic description, it seems likely that UBI + BLKUBI (block driver over UBI) may give similar functionality to what you described but without the performance, because BLKUBI doesn't group 4k writes consecutively. (There are people using that combination in some embedded products in development now.)
However, UBIFS + loop mount, or LogFS + loop mount, might get close in performance. Both of those present a block device which can be used by any filesystem, but underneath translate to a logged tree-structured filesystem. The layers hurt a lot, especially if you loop mount back to user space. We see overhead in raid controllers that hurt performance 30%. Remember that we are dealing with drives that have 40 uS latency on reads and arrays of drives that can hit you with 100 operations a timer tick. UBI is too slow at mounting huge drives, but LogFS is good for that. My reading on LogFS is that it can be very slow with large volumes as well. We mount at around 5 - 20 GB/sec depending on the drive layout. Does your system offer a significant difference from UBIFS/LogFS (containing a single file) + loop mount? I have not seen performance numbers of UBIFS or LogFS published. Our layer on top of a single drive tends to yield ~8000 4K random reads (which is drive dependent) and 4K random writes of around 10,000. Plus you have the benefit of using any file system with any file system feature that you wish. I suspect BTRFS (similar to ZFS) will start getting good at this eventually. It's not aimed at flash right now, but it's a natural direction for a log-tree-structured filesystem to go in when SSDs become more popular. And it has better RAID than drive-level RAID in many ways. Flash is quite "special". If you are really going to exploit it, you have to target everything you do on the write leg for flash. This means designing your code for absolutely zero random writes with allowable exceptions on mount/unmount.
Those platters turned up again mounted on a pod racer. This time right out in the open and not enclosed in a dust resistant area.
In a story so old that most people know about it, they had to use RAST1 (redundant array of stone tablets mode 1 = mirror) because the manager in charge destroyed the first set of tablets, and substantial trouble was gone to replacing the lost data. The data happened to be an ordered list with only ten items on it, so a linear access method was efficient enough for most people. Later the RAST media were stored in a famous enclosure to prevent further damage or unauthorized disclosure. Unfortunately the enclosure was lost, or so we believe, unless it was stored in a large warehouse by the CIA. Use of this data became so commonplace that the items were generally referred to by name and not by number. But this dataset was commonly referred to by the size of the database and the data type of its entries.
At Western Bancorp in El Segundo, we had a vast room filled with 3350's as far as the eye could see. They were attached to the VM/CMS systems (2 195's as I recall). We ran the airline control protocol for use in banking. IBM finance industry stuff...
There I was with the reps from IBM at Boca Raton. I was porting the first word processor for the IBM-PC. We had a spare moment and I looked up and spoke with the manager. I said, "OK I want the game" He said, "You know about the game?". I said, "Yes". He said, "OK ,here it is, and he removed a floppy disk from his jacket pocket and gave it to me. It contained the 8088 dos port of the Adventure game (you know, xyzzy, and plough). He finaly asked how I found out about the game. I said, "No machine reaches this state of development without at least one game." A little social engineering on my part got me an early copy.
They were a must have for trendy nerds. Although they used more power than flash devices, the cute little IBM drives (mine was 1GB) were a triumph of miniaturization and if I hadn't owned one, I might be denying they existed. I bought one to put in my Pocket PC. Unfortunately HP and Microsoft dropped support for the product, but I still have the drive. The state of the art right now for compact flash seems to be 8GB (I just bought one for my Tascam Portastud) and this obsoletes the 1GB device. Unfortunately the device is Type 2 and won't fit in my camera. In contrast though, the Microdrive (1GB ) is substantially more dense than the IBM Hard drive that attached to the 1620 system. It had a capacity to store 100x20K characters. I guest that would be 2 megabytes.
I tried in vain to get people interested in my product idea. A 5-1/4" drawer with zif sockets for three notebook hds. A raid solution for home computers. Kind of like the Drobo but much smaller. I couldn't get anyone excited about what I considered a killer app. All single spindle hard drive based home computers are just waiting to fail. I don't see why raid5 for home computing never took off.
OK I want my mp3's to sound warmer. I am going to buy a Marshall stack for my server. I am committed to the high end. Musicians will prefer tube to solid state storage, I can see that.
The state of the art for CF at the moment seems to be 32GB, not 8.
And yes, the microdrives were triumphs of miniaturization, and they were produced for years. But, the smaller the form factor the drive, the more expensive it was per megabyte or gigabyte.
Due to the cost and the fact that a 1" drive isn't that much smaller than a 1.5", and the spread of a common high speed external interface(USB & Firewire), they lost their niche, became rare enough to get the 'special purpose' price tag, and died against flash.
Even the 1.5" has ceased production if I remember right - they were used in the Ipods, but now Apple(and others) can build their 40GB or so capacity with flash just as cheaply, and often cheaper(no returns due to shock failed HD).
I don't read AC A human right
50 seconds per TB to mount? Tsk, you can do better than that... :-) What are you doing which is taking so much time?
Logfs gets almost exactly the same mount rate, btw. Granted, Logfs is presently limited to 4GiB because linux-mtd is, and it doesn't do RAID etc. or a lot of other things that yours does.
My mistake. I was thinking more about the size of CF that consumer devices have proven compatibility with. As I am sure you realize, just because a widget has a CF intereface doesn't mean it has the hardware support (address lines) to support arbitrary memory sizes). Recently I have purchased several memory devices that didn't end up being compatible with specific devices. My Tascam Portastudio will only support 8GB. My Phone will only support 2GB....