Slashdot Mirror
The State of Solid State Storage
carlmenezes writes "Pretty much every time a faster CPU is released, there are always a few that are marveled by the rate at which CPUs get faster but loathe the sluggish rate that storage evolves. Recognizing the allure of solid state storage, especially to performance-conscious enthusiast users, Gigabyte went about creating the first affordable solid state storage device, and they called it i-RAM. Would you pay $100 for a 4GB Solid State Drive that is up to 6x faster than a WD Raptor?"
Nope. I'd rather wait longer and have more capacity for less money. After all, I use Windows as my primary OS. I'm used to waiting.
Truthfully, though, if the price came down, I'd be interested in this for a Windows install, and then install all my apps and save all my docs to an external IDE.
Mercy was given to me by Christ...I must give the same to others.
The card itself goes for $150, not including any RAM. So add 4 1GB sticks of RAM and you are looking at $500+ for the whole setup. So that is about $125 per GB...ouch!
I remember seeing this sort of thing way back in the DOS days. Battery backed RAM on an ISA card. Product died out because RAM was more expensive than HD.
$150 + (4x$90) = $510 for 4 GB of solid state storage. Definitely not worth it.
"Some days you just can't get rid of a bomb."
The performance numbers Anand came up with on this are a little disappointing, in my view. It's nice, of course, to get a few seconds quicker startup of apps or level loads, but I doubt this is really worth it to most of us at this stage (aside from the coolness factor). Once capacity of these rises enough to make them capable of replacing HDs, though, they might be really nifty in the entertainment/HTPC space due to that silent operation. Basically, an interesting concept, still not quite ready for prime time, but getting a lot closer. Worth a quick read, anyway...
I'd consider buying it if I were building a system that needed some fast write speed... maybe video capture. Be neato if I could get a few and stripe 'em.
If you use this to hold your swap and your main partition, I think the speed improvement would be well worth it! Then buy a 300GB drive for your MP3 collection and all the other junk that that doesn't need such access speed and you are set.
Switching to Linux can be an adventure!
Mmmm, hyper-fast builds that don't depend on the latency of moving parts...
Dewey, what part of this looks like authorities should be involved?
Having disks in parallel doesn't solve the latency problem, only increases the throughput.
:wq
I don't agree. I record music on at least 8 tracks at a time into a single cpu. I NEED higher transfer rates. If it's 4 gigs, thats enough to keep it recording without a drop in an entire days worth of recording. Then I can dump all that data to a slower, larger drive. It may not fit everyone's needs.. but this is PERFECT for me.
** "It's not my job to stand between the people talking to me, and the ones listening to me." -- Pego the Jerk
Nice troll.
It's called "lazy writes", i.e. the OS waits until all the disk buffers are full, or a time limit expires before it writes a buffer to disk. It's a pretty standard operating system optimization - Windows uses it too. "The whole sync() thing" flushes all the buffers and updates the superblock, telling the OS that the file system is "clean". Windows does this also, this is why you see CHKDSK (the Windows version of fsck) running after a rare Windows system crash.
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
It's done because the PCI slot provides continuous power, even when the system is turned off.
What kind of quality are you recording at? At 10 MB per minute (Stereo 44.1 16bit), assuming 8 mono tracks, that's 40 MB a minute, 4000 MB drive (in reality it would probably be less, say 3800) that gives you 100 minutes of recording time. Just over an hour and a half. Now, recording music, it would be more likely, and a good idea, to record at 24 bits and dither later if going to CD, so your record time would be far far less. Just a guess, but probably under 50 minutes. So that gets you 10 5 minute songs. But if you do 2 takes of everything (which is plausible), now you have room for 5 songs. You'll probably do more than 2 takes of quite a few tracks, so...realistically, you'd be able to fit *maybe* two full songs on a 4 gig drive. It doesn't appear that a 4 gig drive would be enough really, unless you were prepared to dump your files quite a few times a day. Doesn't seem worth it. Not to mention that a plain old IDE drive can easily handle 8 tracks, even with a moderate CPU. SS storage isn't there yet for media work, at least not from a cost/performance point of view.
On a sidenote, is anyone going to buy this drive that is 4gb and costs 100 bucks? I don't think it's much use to anyone.
In the era of cheap, throwaway crap, I'm pretty much by myself when I say "I want QUALITY". So yes, I'm planning on buying several of these later today to put them in my main machines in my business. they'll be running our mission-critical cash registers.
Anyone that says this isn't worth it is not very technical in my book.
An affordable 4 GB is fantastic for this kind of thing. Use your imagination:
1. Imagine how fast your system would be installed on a battery-backed up RAM drive.
2. Imagine how fast your system would be with your memory swap file installed on this.
3. Imagine how fast your database server would be with its transaction log installed on this. Hey, throw the tempdb (for SQL Server) on there as well.
4. Many other things.
If you're thinking of this as a standard hard drive to store your DivX movies and MP3 files, you're not thinking right. Solid state drives are miracles that can speed up systems beyond anything you would expect.
I'm a big tall mofo.
I submitted this as a story back on June 4. Since it was rejected (too verbose?), I posted it to my /. journal. My main question to other folks relates to how this
would compare to using a regular ramdisk. The main deficiency with a
ramdisk is that you'd have to reload the contents every time you reboot.
Here's my article, with all its links:
Giga-byte Technology recently came out with a DRAM-based PC card that operates as a SATA hard drive. The product, iRAM, uses power from the motherboard to keep memory active when the system is shut down. During power outages, the product uses a on-board battery to retain memory for up to 90 minutes. The iRAM card is being talked about in the news (InfoWorld, itWorldCanada, engadget, PCWorld, multiplay forum) as a means of booting Windows faster. That is, you install Windows onto the iRAM drive to take advantage of the RAM's faster read-access time. Just hope that you don't lose power for more than 90 minutes.
Is boot time really that important, since many computers are on all the time? A ramdisk might have better uses, perhaps for caching frequently-accessed files such as databases and webservers. Or, if you insist on having faster bootup, instead of putting Windows on the iRAM disk, why not just store the hibernation file there?
I implemented a RAM-based database for an internet tool in 1998 to alleviate the read/write load on my local hard drive. It turned out to be a simple solution for the problem. At the time, it was just a matter of using a DOS-based ramdisk driver (ramdisk.sys). On application startup, it copied the database files to the ramdisk. During operation, everything was read/written to the ramdisk, and periodic backups were made to the physical disk. There are some inherent risks, such as loss of data during a crash since data isn't immediately written to a physical hard drive, so it may not be a great solution for a mission-critical production database. The iRAM product would make this type of database even more stable, in that the risk of loss of data is much less.
That was a while ago, so I thought I'd look into setting up a ramdisk in XP for some amusement. Follows are the results of that search. It seems that the options are relatively sparse beyond the DOS-based driver. A few freeware and commercial packages are available, though. One key factor beyond price is the size limit of ramdisk.
Microsoft's ramdisk offerings since Win2k are limited. Included with the XP OS is a ramdisk sample driver that "provides an example of a minimal driver. Neither the driver nor the sample programs are intended for use in a production environment. Rather, they are intended for educational purposes and as a skeletal version of a driver." Installation isn't simple enough for most users to benefit.
Alternatives include a shareware ramdisk, AR ramdisk (archive link: http://web.archive.org/web/20041011170408/http:/ww w.arsoft-online.de/products/product.php?id=1) (freeware, 2GB limit, discontinued, available for download here), a freeware (64MB limit) and shareware (2GB limit) version here,
On the contrary, I've always been amazed at the rate of price/performance evolution in HD technology.
Consider that in 1982 a 10 MB disk cost something on the order of $3500 while today you can reasonably expect to get an 80 GB disk for $50, that's a drive that has 8000x the storage for 1/70 the price or a price/MB improvement of roughly 420,000x. And, that doesn't take into account the dramatic improvement in reliability and speed (both access and interface) that the newer drives exhibit. Do you think CPUs have kept up with this?
I've heard people predict the end of moving-parts mass storage for years now, but it still seems pretty distant considering the great values we're getting with HD technology.
Power consumption goes up when it is removed from the PCI slot, says the article. If that's so, there is a design fault somewhere - it suggests that there are floating inputs .
Panurge has posted for the last time. Thanks for the positive moderations.
Having disks in parallel doesn't solve the latency problem, only increases the throughput.
Latency comes from three sources:
1) Head latency.
2) Rotational latency.
These are the two sources you have considered. Striping indeed does absolutely nothing to help there.
You forgot the third source of latency:
3) The-disk-is-busy-serving-another-request latency.
Your comment would be true for a primitive OS with a single-threaded I/O method, and/or a RAID system with no command queue.
Given that modern RAID systems are NOT primitive, I/O performance is no longer measured with rotational + head latency vs. throughput, because those measurements no longer make sense.
There are two kinds of performance measurements for modern disk subsystems:
1) MB/sec. (bandwidth) This is what most people think of when they think of throughput.
2) I/O / sec. This measurement is simply the reciprocal of the head+rotational latency in the case of a SINGLE DRIVE. However, in a multi-drive setup, max I/O / sec. increases proportionally with the number of drives, up to a point (eventually you hit the limits for the RAID controller, bandwidth, whatever).
If we measure latency a the time it takes a single drive to physically get the data given a single request, sure, mutiple drives don't help. If we measure latency as the amount of time between when the application asks for the data, and when the disk delivers it, RAID helps quite a bit, beacuse the different I/Os are distributed to multiple disk heads, each of which can contribute it's own I/O handling capacity.
SirWired
Actually I don't know where they even got $100 from because the article says:
"Gigabyte has told us that the initial production run of the i-RAM will only be a quantity of 1000 cards, available in the month of August, at a street price of around $150. "
OH, and did anyone notice the price does not include RAM? So you're paying $150 for a card that can accept up to 4GB not "$100 for a 4GB Solid State Drive ".
That's got to be the most misleading quote ever on a /. article description since u'll spend closer to $500+ for the card and four 1GB DIMMS
my karma will be here long after I'm gone