If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Announcement
Collapse
No announcement yet.
IBM Nanophotonic Scalable to Terabits of data over Fiber Optic
It's super late, and I'm not really seeing the direct implications spelled out in layman's terms so...
Can someone give me the gist of what this means?
You know how we're hitting the upper reaches of Moore's law? The conundrum that electrical engineers are facing is that the process is ever shrinking but you cannot shrink infinitely. This novel solution developed by IBM circumvents this limitation by introducing a whole different way that data will be processed. This also overcomes the hurdle of data transmission. Fiber Optic is currently limited to the hardware of the devices transmitting the data (e.g., your NIC, for example) However, being able to push this ceiling from megabit (which is what currently is at -- even Google's gigabit is actually 30% less than that or just a few hundred megabit short of 1 gbit) to terabits (i.e., more than several terabits) will open up the flood gates to cloud computing which will eliminate the need to have personalized processing.
However, the technology also enables personalized processing (or maybe let's call this consumer cloud) ... what this means is your average Honey Boo Boo will be able to afford Google computing in their own home sans the need for an internet connection. The catch? IBM hasn't decided if it wanted to do that ... so we may have to brace ourselves for the alternate world that Steve Jobs railed against when he launched Apple in the early 80s ... against IBM.
Re: IBM Nanophotonic Scalable to Terabits of data over Fiber Optic
What would they gain by sitting on the technology? All I see that accomplishing is giving competitors time to replicate your results and steal your thunder(read: fistfuls of cash).
Re: IBM Nanophotonic Scalable to Terabits of data over Fiber Optic
A single CMOS-based nanophotonic transceiver is capable of converting data between electric and optical with virtually no latency, handling a data connection of more than 25 gigabits per second. Depending on the application, hundreds of transceivers could be integrated into a single CMOS chip, pushing terabits of data over fiber-optic connections between processors, memory, and storage systems optically over distances ranging from two centimeters to two kilometers.
Okay, this part is VERY interesting. If they can deliver on the "virtually no latency" part and it is cheap enough to put the infrastructure in place then this could really give cloud computing the kick start it needs to really take off.
Re: IBM Nanophotonic Scalable to Terabits of data over Fiber Optic
What is amusing is that IBM makes the CPU for the PS3, Wii and Wii U and 360 (and will more than likely be making the CPU for the next gen Playstation and Xbox).
What would they gain by sitting on the technology? All I see that accomplishing is giving competitors time to replicate your results and steal your thunder(read: fistfuls of cash).
Oh, I meant that they won't license this to anyone else, and monopolize it for themselves, by being the sole provider of: infrastructure, application and support. This could be a multi-trillion dollar industry in itself with IBM being the sole aggregate of any new future extensions of THIS technology tied to THEIR platform.
We're all just hoping they won't be that "DICKHEAD" company ...
What is amusing is that IBM makes the CPU for the PS3, Wii and Wii U and 360 (and will more than likely be making the CPU for the next gen Playstation and Xbox).
IBM is also putting out rival mobile embedded technology that may surpass that of Qualcomm's Adreno/ARM SOCs... and using this nanophotonic technology, it will allow them to essentially put the power of an i7 desktop + nvidia's top GPU into your smartphone with a fraction of the power needs and tremendously lower heat output in just two years ... the mind boggles. Again, assuming IBM actually licenses this tech to the companies in the best position to push the envelope quicker rather than trying to keep this to themselves and put out to market years later ... and history seems to indicate that the latter and not the former will likely happen.
Comment