Research Scientists To Use Network Much Faster Than Internet 50
nickweller writes with this story from the Times about the Pacific Research Platform, an ultra-high-speed fiber-optic research infrastructure that will link together dozens of top research institutions. The National Science Foundation has just awarded a five-year $5 million dollar grant for the project. The story reports:The network is meant to keep pace with the vast acceleration of data collection in fields such as physics, astronomy and genetics. It will not be directly connected to the Internet, but will make it possible to move data at speeds of 10 gigabits to 100 gigabits among 10 University of California campuses and 10 other universities and research institutions in several states, tens or hundreds of times faster than is typical now.
Internet 3? (Score:1)
Had the same thought myself (Score:1)
Throw millions of tax dollars at ISPs to upgrade their systems, what do we get: dead fiber.
Throw millions of tax dollars into creating a faster 'research internet' (internet2), what do we get: dead fiber.
Forgive me if I'm a bit skeptical about throwing more millions of tax dollars to make the next faster internet.
Tiered internet is coming? (Score:1)
No, it's here.. We'll always have it. Some people are still on dialup. I don't know if I like the idea of computer 'islands'. We're just going back to old fashion isolation.
being greedy is gonna cost them (Score:2)
Re: (Score:1)
I think you missed the "how the Internet got built" story.
Re: (Score:3, Insightful)
At a university I worked at in the past, we had routers that could hand way more than 100G of data ten years ago, and could even put that on a single fiber with a multiplexer. It was off the shield equipment. Projects like this use off the shelf equipment, stuff that is already used in major connections and backbones elsewhere. There is no being greedy and saying others can't play. It wasn't cheap (well, sometimes the routers were free from the vendor since they wanted our help testing out new equipment
More porn... (Score:1)
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
> 100gbps
A hell of a lot more than 8k!
Man, I wonder what kind of porn that will be...[looks up to the stars wistfully, eyes like glazed saucers, the wonder of the infinite]
Canada did it first (Score:5, Informative)
Canada did it first [wikipedia.org], And it's country-wide. It can also do speeds as high as 100 Gbit/s but generally operated at 10 Gbit/s.
Re: (Score:1)
you are just adorable. you must have lots of friends
Troll Slashdot if you're lonely (Score:2, Offtopic)
He's actually a perfect example of how well slashdot's moderation system typically works. Oddly enough, he's demonstrating the exact opposite of what he's intending, especially since no one is joining him in his angry little crusade.
Anyhow, back on topic. It seems like we might as well expand the regular internet's capacity to transfer this much data. Streaming video has a practical limitation beyond which there's no point in increasing resolution or fidelity. There's also finite amount of video you'd n
Pointless article (Score:5, Interesting)
1) 10 or 100 gigabits is not a measure of speed.
2) the "current" internet could very well exhibit the same capabilities if it didn't have to carry all the porn streaming left and right for millions of clients. A conventional network connection rated at xxx could run at that rate if you didn't have any sort of congestion, something this new network will likely not suffer because it doesn't have porn (yet). Any dedicated link will give you that. Heck, any 100Gb/s optical channel will give you 100Gb/s to play with.
3) "designed with hardware security features to protect it from the attacks" from the "internet" - by not having a direct connection to the internet in the first place? Fancy words, but it'll do.
4) $5 million to weave a cluster of fibers? Sounds too cheap.
Finally, the article says that "the new network will also serve as a model for future computer networks", but doesn't say anything about protocols, routing, etc. Nothing.
Re: Pointless article (Score:2, Interesting)
It's true. This is all BS. The internet is built on the same tech, it just has to service a lot more than ~20 locations. The internet in Japan has been at these speeds for a while.
This is some Ministry of Truth shit I think. Some kind of brainwashing to make Americans accept when the internet doesn't get any cheaper but does get slower. It's going to happen. They are conditioning us! They must still be salty 'bout that denied merger.
Re: (Score:2)
To be fair there isn't much you can do with 5 million. I imagine that the original internet back in the 70s cost a lot more to put together.
I'll find a good use for it if you can't. Paypal me.
Re: (Score:2)
So what? I had a gigabit home network back in 2001...
Re: (Score:3)
try the sister article linked from soylent news.
they are experimenting with a replacement for tcp/ip optimized for very large packet sizes.
Re: (Score:3)
I guess it means:
a) We rent existing but empty channels/fibers from providers (otherwise 5M would be impossible)
b) We dont connect it to the internet; although they sadly dont mention if they have a private internet (not news) or if they use another protocol to avoid the negative side effects of TCP/IP (little news, unless they show the numbers)
c) If I assume they are talking about 10 to 100 Gbit per second, then it would not be so fast) as far as I understand, single channels in fibers go up to 40GBit/s
Re: (Score:3)
Sounds like Internet 2? (Score:4, Informative)
Sounds a lot like Internet 2
https://en.wikipedia.org/wiki/... [wikipedia.org]
In 2006, Internet2 announced a partnership with Level 3 Communications to launch a brand new nationwide network, boosting its capacity from 10 Gbit/s to 100 Gbit/s. In October, 2007, Internet2 officially retired Abilene and now refers to its new, higher capacity network as the Internet2 Network.
Ohio had it first! (Score:1)
OARnet [oar.net] in Ohio has had this for a while now...
This sounds pathetic (Score:3, Interesting)
Internet 2 was already supposed to be faster than the regular Internet.
This announcement of 10Gbits to 100Gbits is not impressive .. that is a typical server connection these days.
What would be impressive? Multiple terabits or even petabit network, dedicated to the schools, allowing each connected device maximum throughput simutaneously.
Re: (Score:2)
This announcement of 10Gbits to 100Gbits is not impressive .. that is a typical server connection these days.
Heck, in March the first residential 10G/10G Internet connection was delivered here in Norway from Bayonette, source [google.com] via Google translate. They have a 24xGbit hub with 2x10G for expansion and instead give you a direct line. Note that it mistranslates the prices, it's 5999 NOK = $727/month for 10G, 3-400 NOK = $36-50 for 1G so I'd call it mostly a publicity stunt but for a dedicated research network it's peanuts.
Now 100 Gbit is a bit more exotic but I know "The Gathering", a 5000 people data party had a 100
Super Duper Intranet? (Score:5, Insightful)
How fast in Internet??? (Score:5, Insightful)
Internet isn't a speed, it's a concept. The Internet can have connections at any speed.
Re: (Score:1)
But in reality the Internet speed (for a given 2 entities) is the average bandwidth between their 2 connection points as measured over a statistically significant number of transfers.
As others stated thanks to cat videos and porn, this is a lot slower than even modest dedicated hardware.
Re: (Score:2)
The other thing is that the Internet has a lot of overhead.
When it was originally developed, networking was very slow and unreliable, so small packets were picked. As hardware has improved and available bandwidth has grown exponentially, the benefits of larger packet sizes are mostly lost since, for compatibility reasons, everybody continues to use tiny packet sizes in order to avoid dropped/fragmented packets.
Been there... done that... didn't get a T-shirt (Score:2)
Very nice, but hardly new. Both ESnet [es.net] (U.S. DOE research network) and Internet2 [internet2.edu], the national collegiate research network have been running at Nx100G to major research sites and the rest of the Internet for at least two years. They provide Internet service places like CalTech, MIT, the University of Califorrnia, Berkeley Lab and Fermilab. These are full production networks with ESnet already moving vast amounts of data from the LHC to the US for storage and dissemination to many public and private research
Let me guess (Score:2)
Free Market (Score:1)
Why didn't they just wait for a bid from a competitive corporation to build such a network in our wonderful and glorious free market and capitalist America?