Slashdot Mirror
Israeli Startup Claims SSD Breakthrough
Lucas123 writes "Anobit Technologies announced it has come to market with its first solid state drive using a proprietary processor intended to boost reliability in a big way. In addition to the usual hardware-based ECC already present on most non-volatile memory products, the new drive's processor will add an additional layer of error correction, boosting the reliability of consumer-class (multi-level cell) NAND to that of expensive, data center-class (single-level cell) NAND. 'Anobit is the first company to commercialize its signal-processing technology, which uses software in the controller to increase the signal-to-noise ratio, making it possible to continue reading data even as electrical interference increases.' The company claims its processor, which is already being used by other SSD manufacturers, can sustain up to 4TB worth of writes per day for five years, or more than 50,000 program/erase cycles — as contrasted with the 3,000 cycles typically achieved by MLC drives. The company is not revealing pricing yet."
promoting your own company via slashdot much?
If we have to ask how much it costs, we definitely cannot afford it.
Call me when it's 75% cheaper than other "solutions".
Never do what you can in hardware, in software. ...and we can't do this in hardware! :)
So basically, -1 troll/offtopic is really slashdots way of saying "I hate that you thought of something before me."
With Enterprise SSD's (SLC) still in the $100/GB range, we're far away from general acceptance in the datacenter. MLC also has the problem of being slow to write to vs. SLC which is one of the important metrics when considering SSD's to accelerate your classic spindles. SLC's are reliable enough to last for at least 3 years even fully loaded at 3 or 6 Gbps.
I used some Intel X-25-M and Intel X-25-E's in my environment as they are affordable and generally get the highest scores in IOPS and throughput respectively read and write caches and the performance is way under my expectations. The Intel X-25-E's don't work well under heavy loads on LSI controllers (throws errors and SCSI bus resets) while he Intel X-25-M's do work fine. Every other month there is fresh firmware to fix some or another problem and firmware updating is manual labor with a boot CD, not something you can simply schedule at night or do while the system is online so they are what I would call beta-quality. Especially once fully filled the IOPS performance drops from ~3000 IOPS like a brick to ~1000 IOPS which a small set of hard drives can fulfill so the only good thing it's left for is latency.
We'll see what the Vertex 2 EX brings (Sandforce 1500 controller) which has an advertised 50k IOPS although that might be more marketing than anything. I'm still waiting on a decent priced SAS SSD which can actually sustain 5-10000 IOPS by itself even when fully loaded.
Custom electronics and digital signage for your business: www.evcircuits.com
How is this different/better than the sandforce controllers we already have?
I suspect this will eventually bring down the manufacturing costs of Enterprise class drives, rather than making consumer drives "more reliable". I think reliability concerns with current consumer-oriented MLC designs to be overstated.
Anecdotally, my Intel 160GB G2 drive is going on 7 months of usage as a primary drive on a daily used Win7-64 box, and has averaged about 6GB per day of writes over that period (according to Intel's SSD toolbox utility). Given that rate of use over a sustained period (which theoretically means it could last decades, assuming that some as yet undiscovered manufacturing defect doesn't cut it short) combined with the fact that even when SSDs fail, they do so gracefully on the next write operation, I just don't see the need for consumer-oriented drives to sport such fancy reliability tricks.
How can a solid state drive have a "signal to noise ratio"?
It's all digital. Either the voltages are within their valid thresholds or they are not.
Wouldn't you need the world's fastest DSP to "clean up" noisy digital signals and still maintain the type of transfer rates they claim?
There is nothing about this breakthrough that makes any sense. Snake oil?
We already did something very similar to this on the BAIL backup subsystem of the Cassini spacecraft many years ago, and it didn't require a "special" processor.
9/11 Eyewitnesses to Explosive WTC Demolition 1 of 2
The SSD will have a more powerful CPU than the computer.. All it will need is a graphics and audio chip, more RAM and.. oh... nevermind..
Todos mis movimientos están friamente calculados
http://www.imdb.com/title/tt0848228/
The Israelis have Super Star Destroyers?!?
Wait, crap, I just left http://bbs.stardestroyer.net/. My bad.
...for Moore's Law!
Just great, another awesome piece of tech I so desire in my machine that I can't afford. CURSES YOU!
The company is not revealing pricing yet."
They are competing on reliability, so it makes sense the price would be higher.
The fact they are not advertising the price, strongly suggests they do not intend to compete based on price, and price will be high.
Marketing rule #1 is shove all the positive aspects of your product in the customer's face.
Don't talk about the negatives or the disadvantages, if you can avoid it.
In this case the product's not out yet, so they can avoid talking about the high price it will cost at launch :)
Jew lover's what ? That's a possessive apostrophe you twit.
At least you show that you're a bigoted idiot instead of making people work it out for themselves.
So we can have 50.000 instead of 3000 rewrite cycles. That's great. However, I still like the 100.000 to 1.000.000 rewrite cycles of SLC. Actually, SLC is only 50% more expensive to manufacture (per bit) than two-level MLC - I really don't understand why are manufacturers so enamoured with MLC.
"The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
Jew lover's what ? That's a possessive apostrophe you twit. At least you show that you're a bigoted idiot instead of making people work it out for themselves.
Great, Jew lover vs Grammer Nazi. This could be fun in a few minutes
This sounds absolutely no different to how all wear-leveled, error correcting flash controllers work. They all use multiple levels of ECC to decrease the error rate. The 'signal processing' they're doing doesn't sound like anything new.
If there is something new going on here, it's absolutely impossible to decode from the layman's language used in the article. All I hear is "Other vendors use X bits for ECC. We use Y bits and we do it in software instead of hardware.", which is basically just another way of saying "Other vendors have 4 blades, we have 5 blades."
Hello, mirko
Well if they'd stop oiling the spindle with blood and pituitary glands of Palestinian babies, I'd buy from them again. Christ, it's just about impossible to get that shit out of the rug.
Anyone who loves or hates any language, platform, or manufacturer, doesn't know what they're talking about.
Except we've been wear balancing as a standard since the 1990's and we certainly didn't invent it.
Wear leveling was normal for NAND long before that.
What kind of n00b are you?
http://www.google.com/patents?vid=6850443
http://lkml.org/lkml/2005/8/20/95
Thats a shame... i thought they had developed a Super Star Destroyed. Nothing to see here... move along.
From the description (and a lot of guesswork), it sounds a bit like they might have put in a basic RAID system, but using separate memory chips instead of drives. In terms of price vs performance/capacity, RAID has been a good solution, so this might well make sense, IF they don't try to make it out to be some black box filled with magical gold dust, rather than a simple application of existing tech in a new area.
It's just a matter of time before someone would use a stronger ECC. Now each 512-byte sector has extra 16 bytes for ECC checksum, which is enough to recover one bit. Given enough space for the checksum it's possible to recover as much data as needed. There are a lot of implementations in hardware. Every wireless tech designed in the last 20 years uses one, typically amount of extra data is in range 1/6 - 1/2. Hard drives certainly implenent better ECC too.
Now the problem is where to place extra checksums in current NAND chips, but it should be solvable. This problem is about as difficult as implementing wear leveling.
The article says this new technology boosts the number of write cycles from 3000 to 50000. Sounds good, but then again, SLC flash in 1991 supported 1million writes and MLC 100.000 writes. Later consumer grade MLC flash claimed to handle 10000 writes and Micron is selling MLC flash that supports 30000 writes and I recall AMD having MLCs with 100.000 writes. Maybe the 3000 writes MLC is high density & as cheap as possible kind of flash and this new Israeli technology works on that. But unless it is cheaper than Micron flash, I don't see why would anyone use it.
- Raynet --> .
Call me old fashioned but "in a big way" simply doesn't cut it for me. It is language used by pikeys. Or by bullies which are a linguistic or intellect stone's throw away from pikeys.
Do the geek proud and make a bit of an effort when writing. After all, the typical geek reads more than Joe Average -well "he" claims so and I personally do anyway- and hence trains his brain in appreciating well formed sentences.
Besides, there are so many alternatives to "in a big way".
I hadn't the slightest objection to his spending his time planning massacres for the bourgeoisie... (P.G. Wodehouse)
So if Slashdot summaries are stored there before being posted, it will fix all the typos and such?
Two years ago I made a huge mistake letting my gf convince me she needed dead sea dirt for her chronic illness. The result, me and my rented car got nearly STONED to death by Jewish hassidic (sp?) villagers just off the main road. I understand now it's not polite to take pictures of your synagogues shirtless but common, stoning tourists w/o even giving a chance to explain themselves? It did not happen to me but later I heard stories of people killed by Jewish religious students for kissing on the beach. Israel maybe the most democratic and western like country in the ME bit it ain't America or Europe, it's still fucking ME. Just remember it.
US-UK-Israel: The real Axis of Evil
Based on this patent, it looks like they write, read back an analog signal, and then use any deviation from expectations to compute a compensation factor for a second write which is the actual data write. In other words, the first write is used to calibrate the data writes. I assume this calibration is done rarely or the write bandwidth would be 1/3 of a non-calibrating system.
Variations due to process (the cell is smaller or larger than intended) only need to be calibrated once. Variations due to environment such as temperature or supply voltage require more frequent calibration but should hurt bandwidth too much. I assume the MSP determines when calibration is required.
The patent:
http://www.freepatentsonline.com/y2009/0103358.html
Although the vagueness of the marketing material made me skeptical, the patent shows they have a significant contribution and this technology does have value.
Extra ECC data and fancy controller trickery can't get around the fact that the write limit is a limit of the underlying flash, not the controller...
Extra ECC data and fancy controller trickery can't get around the fact that the magnetic media density limit is a limit of the underlying magnetic domains, not the controller...
No wait! Then they invented PRML. Turns out the underlying limit was actually due to engineers lacking vision. All they needed was a new analytic frame of reference. The same deal has happened over and over again with RF spectrum. One man's noise is another man's signal. I just don't know the RF world well enough to cite examples off the top of my head.
That said, there's a long of history of quacks who would like you to believe they invented PRML when they actually haven't.
On gut instinct I'd give this about 3:1 against this having a solid grain of truth, and slightly longer odds against commercialization at significantly better than cost parity compared to other methods of achieving the same end. Even where you find a grain of truth, the product often falls into a niche for one reason or another. Sometimes it's nothing serious, just small things that get refined in the due course of time, which would be great if your massively larger competitors were transfixed with awe.
So many business plans are missing the critical line item:
Transfix deep-pocket competitors with awe while we burnish our new technology to ultimate perfection.
Pity.
Several of the most important corporations in the world have major R&D sites in Israel. You would have to boycott Intel, Microsoft, Sun, Philips, Zoran, IBM, Motorola.. the list goes on..
Besides whats the point? How does you not buying a superior SSD promote peace in any way?
Strong hi-tech companies provide philanthropic aid many times without accordance to government policies. So what your saying doesn't make much sense.
I believe in peace and human rights, but honestly I don't think South Africa is an example of how things should be done.