In honor of the late Neil Armstrong, I'd like to take a look at networking in the final frontier. There are a lot of really interesting things happening in space right now - the landing of Curiosity, the first docking of a privately owned space craft to the International Space Station, space tourism, nanosatellites, and the private Mars One project to colonize Mars are just a few.
This raises questions to my mind that revolve around communication. How will network connections be formed between interplanetary objects? We still have problems with being on the wrong side of celestial bodies, interference, and poor satellite coverage for the current space program. Plus, we don't even have world-wide Internet-connectivity or reliable connections.
What place does traditional, "terrestrial," networking have when there is no end-to-end path. Moreover, how do we bring networking to the stars when we can barely network two WANs together without intermittent connectivity issues?
Delay/Disruption Tolerant Networking
Well, one answer that is being heavily researched is Delay Tolerant Networking (DTN), also known as Disruption Tolerant Networking and, potentially, Delay and Disruption Tolerant Networking.
DTN addresses networking issues in mobile or extreme environments that lack continuous networking connectivity. In conventional networking, if a node can't connect to another node to deliver its packets, the packets are dropped. TCP/IP does not tolerate disruptions between end-to-end paths and times out after a relatively short time period. Packets are dropped and not enough data may arrive at the destination. In the case of interplanetary communications, this can be disastrous since there are frequently times when there is no direct line to the next node.
DTN bundles the data and holds it until it can transmit the information to the next node. Unlike TCP/IP, DTN will store that bundle until it can successfully forward the bundle to another node. This implies extra storage capacity on the nodes so the nodes can contain multiple bundles. It also implies a certain amount of intelligence in DTN nodes to determine not only the best route for the bundle but also the best route without overwhelming the next node with data it cannot store. (Yes, there are many security concerns around storing bundled information.)
So, how will this benefit us poor terrestrials?
The potential benefits of DTN
Well, ignoring the vaguely science-fictional scenario of communicating with deep-sea bases, I can think of quite a number of use cases. Military communications is an easy use case, especially with submarines or with troops deployed to areas without consistent communication lines. I've frequently wished for a data connection in the middle of a hike in the woods so I could figure out what the heck just bit me, so increasing connectivity to the "wilderness" is a fond hope. This could provide more connectivity to developing nations. As the many natural disaster can illustrate, an important use case for everyone is more consistent connectivity during natural disasters.
Assuming this is the future, or a close facsimile thereof, how in the world do you monitor that?