Building the Interplanetary Internet

sighted writes "Internet pioneer Vint Cerf, now a Google VP, is leading a NASA effort to create a permanent network link to Mars within the next two years. As Cerf outlined in a recent talk, the 'InterPlaNet' protocol is designed to handle the delay caused by interplanetary distances. A signal traveling between the Earth and Mars can take up to 20 minutes."

Re:How exactly is this news?

[yincrash]

from the article

"We are working on standardising the protocols to enable spacecraft communicate and share information across the solar system," Cerf said while delivering a talk on the 'Future of the Internet'. "Communication between a rover operator on Earth and a rover on Mars, via a relay orbiter, can't use standard Internet protocols end-to-end. That problem is at the root of a lot of the design work we're doing for the IPN... As part of the NASA Mars mission programme, the project aims to have by 2008 a well-functioning Earth-Mars network."

Re:Ping

[Steve Cox]

From the presentation on IPN I saw a few years back, it appears that you wont be pinging marsbase.com..... they actually fancy adding a couple of levels to get some real TLDs.....

ping marsbase.com.mars.sol

When I saw the .sol in the presentation I was pretty impressed... theres a little bit of future proofing in that one....

Steve.

What is the maximum latency for communication?

[jmagar.com]

I wonder what the acceptable max latency could be for 2 way communications. We have become quite used to having near immediate mode communications, and computer networks are possibly dependant on it. At what duration in time does distributed computing fall down? What is the maximum time to live on a TCP packet?

I'll be keeping an eye on this to see how they address these sorts of issues. Also, does this not relate to RFC 1149 [ietf.org]? Certainly the latency issue is common.

Re:How exactly is this news?

[skoaldipper]

I think this is news now since one node of the future IPN is scheduled to be functional by 2008. The article doesn't specify in what way though; albeit a new orbiting satellite (gateway) around Mars, or just reprogramming some existing Mars mission device already deployed (to test the IPN protocol).

I like pretty pictures and diagrams. So, I found a good presentation [spaceref.com] by Cerf back in 2000 which outlined these challenges and why [spaceref.com] we need the IPN.

Re:Ping

[$RANDOMLUSER]

ping marsbase.com.mars.sol.milky
Now it's really future-proof

What a fun project!

[FuzzyDaddy]

I had a friend who was doing some consulting for a company that wanted to offer satellite based internet connectivity. When they first tried out the system, things took forever to download, despite the fact they had many Meg of bandwidth. Each picture that loaded involved a separate TCP/IP connection, which takes several back and forth messages to establish - which was sluggish because of the latency going to the geosynchronous satellite. (This was several years ago, and all the vendors have very sophisticated understanding of the issues).

With a twenty minute delay, the standard practice of resending dropped packets becomes more prohibitive (the send/NAK/resend would take an hour!), so you'd have to make the encoding redundant enough so that most errors could be recovered by the receiver - without doubling the bandwidth. Oh, it would be fun!

Ok, I'll go back to writing documentation now. >sigh

Don't worry

[pato101]

They are going to use MAT (Mars Address Translation)

talking without delays using quantum entanglement

[PermanentMarker]

Just an idea why not used entangled atoms to bypass this distant problem?.
As far as i know there is no limit on distance, changes in one atom happens at the same time on the other atom altough they are on different locations. Thats a quantum physic property

But i'm not sure if information can be passed trough this method (wel hack thats worth investigation)

Re:So P2P now means planet-2-planet ;)

[Loconut1389]

But if the US gets there first, will we claim the whole planet? Its not like there's contintents.. So- will we claim all of mars, or not? If so- then it might be considered a territory of the US and MPAA still would have reign/rights.

Re:Open protocol

[everphilski]

You validated my post. Most, not all, use CCSDS. And while you can do IP over CCSDS (or at least they have a redbook describing it) it hasn't been done, so far as I can tell. And layering a protocol within a protocol isn't the best solution out there, IMO, you might as well do it right and come up with an IP for space.

Re:How exactly is this news?

[peragrin]

I am actually surprised no one has done this before. The Nasa people can figure out specifics, but a number of comm satellites in mars orbit, plus an number in ours to act as relays to mission headquarters.
  Your talking about maybe a dozen total communication relays, and then every probe sent to mars would only need to be able to reach orbit saving lot's of power for other tools and test equipment.

Charge the ESA, russia, or anyone else a bit of cash for relay use, and help pay for it.

Just hope they don't lose their wi-fi connection and you have to go their to reboot the machine.