Sign in / up

back to article This storage upstart knows its technical onions: Symbolic IO

Symbolic IO is a startup that claims its storage and compute technology can run database queries 60 times or more faster than other systems, and offers ”limitless enterprise storage.” It was founded in 2012 by Brian Ignomirello, a former CTO of Hewlett Packard’s Global storage unit, and secured $1.4m in debt financing in 2013 …

COMMENTS

Post your comment
House rules Send corrections
Add to 'My topics' unstar *
  1. Michael Wojcik Silver badge

    April 1 is back thataway

    it seems it should be taken seriously. Here’s a tweet from Micron exec Rob Peglar

    Argument by tweet? This has to be the weakest appeal to authority that I've seen in days.

    Not that it would be persuasive anyway. Even if we believe Peglar, and consider him an authority (rather dubious, for technical questions), all he says is that he's "seen it". That's pretty much the weakest form of evidence after hearsay.

    The patent claims quoted in the article sound like plain Huffman compression, or possibly Lempel-Ziv. Unless the patent goes on to say something far more radical, it's hard to see that there's anything at all interesting in it.

    It somehow stores data in a shrunken form; deduplication is not mentioned and nor is compression.

    That sentence is nonsense. Deduplication is a form of compression. Any "shrunken form" is a form of compression. That's what compression means.

    At any rate, the claim made by the article headline is most definitely not supported by the body. There might be something to Symbolic IO, but everything described in the article sounds like snake oil.

    3 0 Reply
  2. Anonymous Coward
    Anonymous Coward

    I had someone ask me about this recently, my initial reaction was snake-oil too, having said that, there might be something to creating a wide dictionary compression algorithm that segments and analyses data at a bit, rather than byte level (I'm pretty sure LZ and Huffman both work with 8-bit bytes as the smallest unit) .. though you'd need an absolute boatload of RAM to make it work effectively.

    If you can get data reduction good enough enough to make storing all this stuff in fast DRAM (using NVDIMMs maybe), economical, and combine that with FPGA's that can run some interesting algorithms in about 200 nanoseconds, then sub microsecond response times are theoretically possible

    Thats a LOT of "in theory" and "maybe you could", I suspect that there is something there, but its probably applicable to a few edge cases where the nature of the workload warrants the hardware centric approach.

    Given the hyperbole though, I suspect that their marketing guy is the same one who created simplistic.io

    1 0 Reply

POST COMMENT House rules

Not a member of The Register? Create a new account here.

Forgotten password ?

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon