Skip to content

If Memory Serves Me Correctly…

February 21, 2012

If you’re into the geek these days, you’ve probably been following NAND technology with some interest.  Persistent memory storage made the iPxx products what they are today.  It’s tough to find a consumer device that isn’t somehow incorporating some variant to store user data.  Heck, Costco is selling it, and getting a flash drive is everyone’s favorite give-away even in the most esoteric of places.  I should add that the company I work for has a thriving memory business, and that I’m not using any inside info in my opinions here.

In fact, I want to spend more time thinking about possibilities than about reality, but first, for the reality, let’s check out The Register:

Microsoft and University of California San Diego researchers have said flash has a bleak future because smaller and more densely packed circuits on the chips’ silicon will make it too slow and unreliable. Enterprise flash cost/bit will stagnate and the cutting edge that is flash will become a blunted blade.

The boffins presented their findings in a paper [PDF] at the Usenix FAST’12 conference in San Jose. They wanted to see how process size shrinkage and the increase in data density – expressed as the number of bits per cell – affected the performance of the NAND and, perhaps, lessened its differential from hard disk drives in terms of performance, reliability and endurance.

I’d like to take this moment to use my own memory (not some silicon storage option) to find some of the embarrassing times that I doubted technology only to watch it smoke my expectations.  If I start recounting them, I won’t be able to finish.  Just one: In the mid ’90s I made a prediction based on some really good data that memory capacity increases on the PC would have to slow down to match the declining prices of the market itself.  I had lots of good data, and watched it all evaporate as DRAM vendors dropped their prices in order to keep their factories busy.  I should have known.

By the way, I agree with nearly every argument made in the PDF.  I also know that there’s  a set of geeks out there working not on lengthening the line of existing products, but finding new lines that will answer the deficiencies.  Every time I begin to doubt the extent of what technology can do, I get surprised, so at this point, I’ve decided that even if I can’t find that particular pony, someone else will.  I’ll stick with sunny optimism in the technology field.

And if you don’t believe me, believe my friend L.  I’ve seen this story a couple times, and it doesn’t matter if it’s true, becuase you know some variant of it is true hundreds of times a year wherever there are smart and lazy people to make it true.

Advertisements
5 Comments leave one →
  1. February 21, 2012 11:30 pm

    Fry’s currently has 32G USB stick drives for about $20, and 128G ones for $349. That price differential suggests that the market has a little room to give. Plus the typical flash drive is not large enough to accommodate the typical drive size.

    So the needs are there, even if the technology seems to be facing diminishing returns. The investment returns will likely apply the needed pressure. We shall see.

    Have you been following the “Gleickgate” Heartland saga?

    ===|==============/ Keith DeHavelle

    • February 22, 2012 8:19 am

      The real issue on NAND comes with business reliability and using it for mission critical applications. But even with that, there are checks that still make the performance and cost differential too meaningful to pass. Consumers will likely continue ot revolutionize the way they live due to technologies like this.

      I made a comment above on Dr. Gleick. I might expand it a bit more when I’m less cranky and cramped.

      • February 22, 2012 10:23 am

        The article that you linked suggests that the Koch brothers “donated $25,000 to the group”— and forgets to mention that this was for health care, nothing to do with climate.

        In other words, they’re pretending to believe the fake memo over the actual real records that were published.

        ===|==============/ Keith DeHavelle

        • February 22, 2012 2:27 pm

          I’d more attribute that to press laziness than true conspiracy.

          • February 22, 2012 2:47 pm

            Perhaps. But the Koch brothers are hated by the Left, burned in effigy, emblazoned across rallying posters et cetera in this publication’s locality. The only reason to even mention them is because the one faked document suggested that they were contributors to Heartland’s climate work, when the real documents taken from Heartland by fraud (and published from the same source) showed that he Koch contribution on climate issues was precisely zero.

            So the name gets trotted out to be hissed at, despite how unwarranted that act is (and how dishonest it is to repeat such a statement).

            I never suggested conspiracy, incidentally; the alternative is malicious intent, which is another matter whether or not it was coordinated with the malicious intent of many others holding up the fraudster as a hero today.

            I frankly don’t think that a conspiracy, per se, is necessary: They share the Cause, thus arrive at similar places. And in fact if you stray from the Cause, as the NYT’s Revkin did in daring to criticize Gleick, you are very quickly attacked as a “corporate shill.”

            I’ve looked at a lot of coverage of this in the past several days; dozens of articles and posts, and thousands of comments. It’s fairly disturbing, and I just put up a bit of a recap last night.

            ===|==============/ Keith DeHavelle

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: