solid state drive SSD latest news

x-posted to the laptops group -
>>>>>>>> RnR wrote:
>>>>>>>>>> http://www.pcmag.com/article2/0,2817,2337194,00.asp
>>>>>>>>> 512GB 2.5" SSD drive? That ought to hold you for a while. Bet it
>>>>>>>>> has a big price tag, too.
>>>>>>>>>
>>>>>>>>> I'm waiting for some reliability studies on SSD. Flash has
>>>>>>>>> the interesting property of failing to hold a charge after so
>>>>>>>>> many writes. When the failure point gets to around a million
>>>>>>>>> writes, I think I will be sold. But I'm not holding my
>>>>>>>>> breath waiting for it. "MKiyoshi Kobayashi, vice president of
>>>>>>>>> Toshiba Corporation's semiconductor company..." Wasn't he
>>>>>>>>> the guy talked about by the Kevin Spacey character in "The
>>>>>>>>> Usual Suspects"? ... Ben Myers
>>>>>>>> Thanks Ben. I'm not up to date with SSDs so your post helps
>>>>>>>> me.... not that I'm going to rush out to buy one but I like to
>>>>>>>> be in the know :) .
>>>>>>> I use them all of the time and I never want to go back to
>>>>>>> mechanical hard drives. As you have to be so careful of hard
>>>>>>> drive head crashes. With SSDs, this problem disappears. The
>>>>>>> mean time between failures (MTBF) with SSD is 228 years. But if
>>>>>>> you want to wear one out, you have to rewrite the whole SSD 24
>>>>>>> times a day for the next 11 years to actually do so. ;)
>>>>>> Please cite the source of these reliability numbers. Thanks...
>>>>>> Ben Myers
>>>>>
>>>>> Hi Ben! No problem! :)
>>>>>
>>>>> http://blogs.zdnet.com/Apple/?p=901
>>>>>
>>>> I would assume that the numbers cited in the ZDnet blog come from
>>>> manufacturers' theoretical numbers. Like most modern faux
>>>> journalists, the blogger does not explain where he got the numbers,
>>>> so we're back at the same conundrum: What is the ORIGINAL source of
>>>> these reliability numbers.
>>>>
>>>> I'd still like to see some real world
>>>> beat-the-hack-out-of-the-drive-for-a few-weeks tests... Ben Myers
>>>
>>> Might as well go to the source for info:
>>>
>>> http://download.intel.com/design/fla...-datasheet.pdf
>>>
>>>
>>> According to Intel specs on its X-25E SATA 32GB SSD:
>>>
>>> weight 80 grams (approx 4 ounces)
>>> MTBF 2,000,000 hours (228 years)
>>> power consumption 2.4w active, 0.06w idle
>>> 75 uSec read latency, 85uSec write latency
>>> 250Mbyte/sec peak sequential read speed
>>> 170Mbyte/sec peak sequential write speed
>>>
>>> Impressive combination of performance and power consumption. Intel's
>>> engineers have been pretty good (not infallible) in working
>>> this stuff out, but I'll reiterate that I'd like to see real world
>>> tests just to get more comfort.
>>>
>>> The $700+ street price tag implies that there is a lot more than
>>> 32GB of flash to cover failures along the way, and the circuitry to
>>> balance use of flash. This is still more than I want to spend, but
>>> prices will drop. Early adopters and technology freaks will pay
>>> the price. ... Ben Myers
>>>
>>>
>>>
>>>
>>
>> Typical Intel NAND Flash specs (from Intel SD74 NAND Flash Memory
>> spec): Data retention: 10 yrs.
>> Endurance: 100,000 PROGRAM/ERASE cycles.
>>
>> Storing the kids' pics or a spreadsheet every now and then is almost
>> risk free. However, a runaway program hammering on this "disk" would
>> not be good.
>>
>> Frank
>
> And, Frank, you have hit the nail on the head as to why I remain
> concerned about the long-term reliability of SSDs. It takes a lot of
> hardware logic and a serious amount of extra flash memory to go from
> 100,000 PROGRAM/ERASE cycles to an MTBF of 2,000,000 hours (228
> years). Maybe flash with far more than 100,000 cycles? Or maybe
> some smoke and mirrors? Engineering design by Bernard Madoff? I'd
> like somebody to explain with a good solid coherent explanation and
> real math computations.
>
> We will all stay tuned for real world stories of how these SSDs hold
> up under regular use... Ben Myers

Micron and Intel have developed some flash memory with a 1 million write
cycle life.

>>> Frank
>> And, Frank, you have hit the nail on the head as to why I remain
>> concerned about the long-term reliability of SSDs. It takes a lot of
>> hardware logic and a serious amount of extra flash memory to go from
>> 100,000 PROGRAM/ERASE cycles to an MTBF of 2,000,000 hours (228
>> years). Maybe flash with far more than 100,000 cycles? Or maybe
>> some smoke and mirrors? Engineering design by Bernard Madoff? I'd
>> like somebody to explain with a good solid coherent explanation and
>> real math computations.
>>
>> We will all stay tuned for real world stories of how these SSDs hold
>> up under regular use... Ben Myers
>
>
** Posted from http://www.teranews.com **

Brand new devices @ 32Gb just being sampled. Probably be 3 or more
quarters before these are in retail SSDs. But these will make a big
difference, since wear-leveling is based on the capacity * write-limit
product.

Frank

Barry Watzman wrote:
> Micron and Intel have developed some flash memory with a 1 million write
> cycle life.
>
>>>> Frank
>>> And, Frank, you have hit the nail on the head as to why I remain
>>> concerned about the long-term reliability of SSDs. It takes a lot of
>>> hardware logic and a serious amount of extra flash memory to go from
>>> 100,000 PROGRAM/ERASE cycles to an MTBF of 2,000,000 hours (228
>>> years). Maybe flash with far more than 100,000 cycles? Or maybe
>>> some smoke and mirrors? Engineering design by Bernard Madoff? I'd
>>> like somebody to explain with a good solid coherent explanation and
>>> real math computations.
>>>
>>> We will all stay tuned for real world stories of how these SSDs hold
>>> up under regular use... Ben Myers
>>
>>
> ** Posted from http://www.teranews.com **

In comp.sys.ibm.pc.hardware.storage Cmplx80 <Cmplx80@karma.com> wrote:
> Brand new devices @ 32Gb just being sampled. Probably be 3 or more
> quarters before these are in retail SSDs.

Excuse me? I just bought one 4 days ago and it is in the mail...

Arno

> But these will make a big
> difference, since wear-leveling is based on the capacity * write-limit
> product.

> Frank

> Barry Watzman wrote:
>> Micron and Intel have developed some flash memory with a 1 million write
>> cycle life.
>>
>>>>> Frank
>>>> And, Frank, you have hit the nail on the head as to why I remain
>>>> concerned about the long-term reliability of SSDs. It takes a lot of
>>>> hardware logic and a serious amount of extra flash memory to go from
>>>> 100,000 PROGRAM/ERASE cycles to an MTBF of 2,000,000 hours (228
>>>> years). Maybe flash with far more than 100,000 cycles? Or maybe
>>>> some smoke and mirrors? Engineering design by Bernard Madoff? I'd
>>>> like somebody to explain with a good solid coherent explanation and
>>>> real math computations.
>>>>
>>>> We will all stay tuned for real world stories of how these SSDs hold
>>>> up under regular use... Ben Myers
>>>
>>>
>> ** Posted from http://www.teranews.com **

Hey, that's great.
How about dropping a hint about what you bought, exactly: An SSD, a
Flash chip?

Frank

Arno Wagner wrote:
> In comp.sys.ibm.pc.hardware.storage Cmplx80 <Cmplx80@karma.com> wrote:
>> Brand new devices @ 32Gb just being sampled. Probably be 3 or more
>> quarters before these are in retail SSDs.
>
> Excuse me? I just bought one 4 days ago and it is in the mail...
>
> Arno
>
>> But these will make a big
>> difference, since wear-leveling is based on the capacity * write-limit
>> product.
>
>> Frank
>
>> Barry Watzman wrote:
>>> Micron and Intel have developed some flash memory with a 1 million write
>>> cycle life.
>>>
>>>>>> Frank
>>>>> And, Frank, you have hit the nail on the head as to why I remain
>>>>> concerned about the long-term reliability of SSDs. It takes a lot of
>>>>> hardware logic and a serious amount of extra flash memory to go from
>>>>> 100,000 PROGRAM/ERASE cycles to an MTBF of 2,000,000 hours (228
>>>>> years). Maybe flash with far more than 100,000 cycles? Or maybe
>>>>> some smoke and mirrors? Engineering design by Bernard Madoff? I'd
>>>>> like somebody to explain with a good solid coherent explanation and
>>>>> real math computations.
>>>>>
>>>>> We will all stay tuned for real world stories of how these SSDs hold
>>>>> up under regular use... Ben Myers
>>>>
>>> ** Posted from http://www.teranews.com **