• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Internet Replacement Coming Soon?

Terry Olaes

I Used to be the [H] News Guy
2FA
Joined
Nov 27, 2006
Messages
4,646
Cern, the particle physics center that created the web, has developed a grid-based system that is reportedly “10,000 times faster than broadband connections.” Cern said that this summer will feature a “red button” day where they will switch on the Large Hadron Collider (LHC), a particle accelerator that will probe the universe’s origins, and the grid at the same time to capture the data the LHC generates.

Computers on the grid can also transmit data at lightning speed. This will allow researchers facing heavy processing tasks to call on the assistance of thousands of other computers around the world. The aim is to eliminate the dreaded “frozen screen” experienced by internet users who ask their machine to handle too much information.
 
Imagine one day every computer is connected to each other at light speed and could transmit terabytes of data a second.... all the computers could constantly talk and call on each other's processing capabilities
 
Imagine one day every computer is connected to each other at light speed and could transmit terabytes of data a second.... all the computers could constantly talk and call on each other's processing capabilities

And that's when SkyNet becomes self-aware. ;)
 
to be honest when you use it on a daily basis it doesn't seem all that special...

The connections may exist but the grid middleware still has a long way to go before it is 100% reliable, heck, even if 90% of my jobs worked on a regular basis I'd be happy over the current 75-85%.

Still, I'm just being picky, we still have a few months to sort out all the kinks, also, the LHC may turn on in July/August, but the experiments (ATLAS, CMS, LHCb, ALICE) won't be recording anything useful for 2-4 months after that, there's a crap load of calibration and understanding of the beam and detectors to go on first.

before you ask, yes I currently work at Cern and I've lurked on the [H] forums for a few years now.
 
Al Gore is pissed!
2004-al-gore.jpg
 
Man...all that pr0n is gonna spread like crazy :p

But yet again, until hard drive technology (esp SSD's) improve and stabilize, this wouldn't help most consumers.
 
Man...all that pr0n is gonna spread like crazy :p

But yet again, until hard drive technology (esp SSD's) improve and stabilize, this wouldn't help most consumers.

Seriously. "Hard Drives" are pretty far behind compared to other computer components.
 
IMHO this technology needs to stay within the research and academia field. There is no purposeful reason for any consumer interaction to come close to this capability. International business may also benefit from something like this but only if the business community built it themselves. The future well being of mankind could greatly benefit from this type of power. No good would come from introducing it to the commercial market. I can live without pOrn at light speed and Team Fortress 2 pings in the low to high teens for the time being.
 
Why exactly are they calling this an internet replacement?

The internet isn't the cables the connect the computers. It doesn't matter if the backbone is running copper or fiber or some new material. It all does the same thing. They could replace the entire backbone of all networks with something new but it is still the same old internet.

As much as people like to talk about a "new" internet, it would be kinda hard to come up with a new replacement overnight that could change the entire way that people communicate with other computers over a connection. Just look at IPv6. How long as that been around yet they haven't moved everything to that over night. Moveing equipment to faster connections overnight, sure that can happen. Although that isn't a replacement for the internet, that is just a speed change. Just like when your ISP purchases a new connection to provide more bandwidth.
 
Why exactly are they calling this an internet replacement?

The internet isn't the cables the connect the computers. It doesn't matter if the backbone is running copper or fiber or some new material. It all does the same thing. They could replace the entire backbone of all networks with something new but it is still the same old internet.
Agreed. The article is so ludicrously light on facts and so heavy on speculation it reminds me of <your favorite candidate>'s speeches.
As much as people like to talk about a "new" internet, it would be kinda hard to come up with a new replacement overnight that could change the entire way that people communicate with other computers over a connection. Just look at IPv6. How long as that been around yet they haven't moved everything to that over night.
ipv6 isn't backwards compatible, and the scheme they were using was declared a success and closed. This doesn't make it easy for Average Joe to migrate, and it's not any better for Average ISP. Thus, stagnation, or at best little pools of ipv6 disconnected from each other.

How to fix? Raise the price of public ipv4 IPs, say, fivefold, starting six months from now. Drop a couple of the mandatory parts: ipsec, for example, would be a prime target. ISPs will figure it out and create 6to4 gateways and put their customers behind them, then get rid of the v4 stuff in the middle.
 
donno the article didnt seem like it was about a fast internet but it was about using multiple computer like fold @ home to figure out the orgin of the universe.
 
I thought it was a red letter day. I'd ask Gordon Freeman, but it appears he's running a bit late for the testing chamber. Of course, I'd rather he not be anywhere near that thing when it's switched on :)
 
And from the resident supercomputing person, a resounding "meh."

This is nothing more than re-announcing the same crap that's being rapidly deployed - albeit on smaller scale - all over the place. Theoretically, DWDM does 400Gb/sec, sure. On a piece of gear that costs more than every poster in this thread will likely every make in their lifetime. It's one-off, highly specialized equipment. Even 'commodity' DWDM equipment starts at about $100K and that'll only get you a couple gigabits on single-strand. CERN's "revolution" really is packing 12x IB frames into DWDM carrier; InfiniBand is not new. The only new really, is the transport. And given how picky InfiniBand is over several meters (I can honestly say it's worse than CrayLink was about grounding,) it actually severely limits single area capacity. They can only get X number of raw IB - I'd guesstimate somewhere at 100 or so - before being forced onto very expensive fiber.

That said, IB isn't internet. It's not even IP. Yeah, InfiniBand frames can carry IP payload. Shedload of bandwidth too. But IP can't carry an RDMA request, is too slow and processor intensive to keep latency usable. I'd just about kill for IB over DWDM, just because that lets me take a CG-256 out from the current spec limit of 256 nodes by 4 rings to something like 1024 nodes by 16 rings, guesstimating. (There's bell curves involved.)

Call me when I can get two strand DWDM for home for DSL prices. And yeah, the article sucks.
 
The Internet is a combination of hardware, software, cables, raw data, and people. It's not something that can become obsolete, replaced, or destroyed. It evolves. In reality the 'internet replacement' they're talking about won't affect the average Joe aside from trickled down technology in the next few years/decades.
 
cool....imagine viruses and malware written just seconds ago making it at light speed to your desktop!
 
Cool.

Can't wait to read about how Comcast figures out a way to cap the bandwidth on it to prevent users from downloading anything it doesn't like.
 
And that's when SkyNet becomes self-aware. ;)

Also maybe most people will have a lower/closer ping to eachother while gaming, I won't have to worry about having my usual ping get jacked up when/if I move to another state(in USA). Heck for right now Fiber Optic would be the bomb, I've got 1.5Mb DSL atm--can't believe I had 56K dialup :eek: SO SLOW.
 
ipv6 isn't backwards compatible, and the scheme they were using was declared a success and closed. This doesn't make it easy for Average Joe to migrate, and it's not any better for Average ISP. Thus, stagnation, or at best little pools of ipv6 disconnected from each other.

How to fix? Raise the price of public ipv4 IPs, say, fivefold, starting six months from now. Drop a couple of the mandatory parts: ipsec, for example, would be a prime target. ISPs will figure it out and create 6to4 gateways and put their customers behind them, then get rid of the v4 stuff in the middle.

There are 6to4 gateways already. Very few places do run ipv6 at their location then talk to the rest of the world using ipv4. But that is kinda what I was trying to point out, the converison to ipv6 isn't an easy one and thus has not taken off very well. Just like trying to comepletely remove what we know as the internet. Then come up with new protocols, new naming schemes, new programs, new OSs, new everything to replace how we do it now which is what some of the internet replacement articles talk about. When looked at from that standpoint, ipv6 full deployement is far more likely.
 
want to see what they're really using that beam for?

check this out

its really just a glorified pigeon gun!
 
We'll survive the Black Hole experiment, but hundreds died. As a result the economy in Europe collapsed, which caused instability in the region which eventually let the the nuclear war which destroyed the world in December 2012.
 
Why exactly are they calling this an internet replacement?

The internet isn't the cables the connect the computers. It doesn't matter if the backbone is running copper or fiber or some new material. It all does the same thing. They could replace the entire backbone of all networks with something new but it is still the same old internet.

As much as people like to talk about a "new" internet, it would be kinda hard to come up with a new replacement overnight that could change the entire way that people communicate with other computers over a connection. Just look at IPv6. How long as that been around yet they haven't moved everything to that over night. Moveing equipment to faster connections overnight, sure that can happen. Although that isn't a replacement for the internet, that is just a speed change. Just like when your ISP purchases a new connection to provide more bandwidth.


Agreed. Even if they use alien communication techniques, it's still the internet, it's just another meduim.
 
And from the resident supercomputing person, a resounding "meh."

This is nothing more than re-announcing the same crap that's being rapidly deployed - albeit on smaller scale - all over the place. Theoretically, DWDM does 400Gb/sec, sure. On a piece of gear that costs more than every poster in this thread will likely every make in their lifetime. It's one-off, highly specialized equipment. Even 'commodity' DWDM equipment starts at about $100K and that'll only get you a couple gigabits on single-strand. CERN's "revolution" really is packing 12x IB frames into DWDM carrier; InfiniBand is not new. The only new really, is the transport. And given how picky InfiniBand is over several meters (I can honestly say it's worse than CrayLink was about grounding,) it actually severely limits single area capacity. They can only get X number of raw IB - I'd guesstimate somewhere at 100 or so - before being forced onto very expensive fiber.

That said, IB isn't internet. It's not even IP. Yeah, InfiniBand frames can carry IP payload. Shedload of bandwidth too. But IP can't carry an RDMA request, is too slow and processor intensive to keep latency usable. I'd just about kill for IB over DWDM, just because that lets me take a CG-256 out from the current spec limit of 256 nodes by 4 rings to something like 1024 nodes by 16 rings, guesstimating. (There's bell curves involved.)

Call me when I can get two strand DWDM for home for DSL prices. And yeah, the article sucks.
Thank you. :)
 
Back
Top