Forgot your password?
typodupeerror
Mars Networking Space

Building an IT Infrastructure Around Mars 121

Posted by kdawson
from the set-the-timeouts-really-long dept.
bfwebster writes "Space.com has an article talking about the efforts to observe the arrival of the Phoenix lander on Mars this coming May using current Mars orbiters. This community will likely be intrigued to see the ways in which NASA is using existing landers and orbiters to prepare for, and then monitor, that landing. This includes using the landers Spirit and Opportunity to simulate transmissions from Phoenix as a testing procedure in advance of the actual landing; using the Odyssey orbiter as a high-speed data transmission link from Phoenix to Earth during the landing; and using the Mars Reconnaissance Orbiter and Mars Express orbiter as backup data stores for Phoenix data transmissions during the descent. How long until we get a terabyte solid-state dataserver (running IPv6, natch) in orbit around Mars?"
This discussion has been archived. No new comments can be posted.

Building an IT Infrastructure Around Mars

Comments Filter:
  • Hmm (Score:5, Funny)

    by moogied (1175879) on Monday March 03, 2008 @09:05PM (#22630478)
    I would guess the line leasing fee's to be out of this world. *cough*
    • Re: (Score:2, Funny)

      by visionlink (715186)
      never underestimate the bandwidth of a station wagon loaded with tapes hurtling thru space.
      • by WeblionX (675030)
        "Sir, we're picking up an outline of a station wagon."
        "Don't bother firing, we're looking for a Winnebago."
    • Not only that, the ping times will be a tad longer than normal. I know I wouldn't want to be up there playing WoW, that's for sure!
    • by Whiteox (919863)
      The lag would be terrible!
      Imagine trying to do chat.
  • This is too much! Someone should do something with Australia's data allowance per month first. 2GB a month for $29 is just ridiculous.
  • Talk about long distance. Maybe they should go wireless. Anyway, this idea is plain cool to me; an astronomy-lover fanatic. I have NO idea what they would do with it, but it'd still be cool to have an IT server around our planet just around the block in the Sol System neighborhood (wow, that sounds cheesy. Oh well, I always say 'Corn and Cheese makes the world go 'round').
  • Does TCP/IP even support a 30 minute latency?
    • by sharkey (16670)
      Yes. [wikipedia.org]
    • by QuantumG (50515) *
      Of course it does. Time To Live is a count of hops not time.

      • That's what I thought too, but from RFC 791, http://www.faqs.org/rfcs/rfc791.html [faqs.org]

        > The time is measured in
        > units of seconds (i.e. the value 1 means one second). Thus, the
        > maximum time to live is 255 seconds or 4.25 minutes. Since every
        > module that processes a datagram must decrease the TTL by at least
        > one even if it process the datagram in less than a second, the TTL
        > must be thought of only as an upper bound on the time a datagram may
        > exist.
        • by QuantumG (50515) *
          What is hard to understand here? There are no routers between Mars and Earth.. therefore the number of "modules" that process a datagram on the way from Mars to Earth is zero.

          • I thought the text spoke for itself: according to that document, the TTL is not the number of hops, but rather the number of seconds a packet has before it expires, minus 1 for every router it goes through. So in practice it appears as the number of hops so long as the trip time is less than one second. I don't know how that would extend to interplanetary transmission (or which side would be responsible for timing the link delay and decrementing the field), I just wanted to debunk the TTL = hops argument.
            • by QuantumG (50515) *
              No, the text you quoted is saying that *effectively* the TTL is a number of seconds argument, as there is no way for a router to indicate that it took less than a second to do a transfer.. so if your packet goes through 255 routers in under a second for each, it will die just as it will if it goes through 255 routers in over a second for each.
              • > "No, the text you quoted is saying that *effectively* the TTL is a number of seconds argument"

                I think you meant to say "number of hops", in which case we're not in disagreement about that so long as you use the qualifier "effectively". If you did mean "number of seconds" then I don't understand your point, as you seem to be contradicting your original statement. Mine was that the TTL cannot be regarded in a general technical sense as a "count of hops not time", but is rather a mixture of both.

                The quest
                • by QuantumG (50515) *
                  When you decide how much the TTL value should be decremented by for a particular link you are not, in any way, restricted. You can decrement it by just one (this is what most routers do) or you can decrement it by 20.. what you can't do is decrement it by less than one. So if you setup your network to decrement by one for every second that a packet spends traveling over a link, as the RFC recommends, then you will have a problem for links under 299,792 km long and over 76,446,960 km long as you can't decr
                • by Jesrad (716567)
                  In our present configuration, there are two routers: one at Earth, and one at Mars.

                  Each of them would be discarding the packets as soon as it received them for being expired, unless their TCP handling was altered for the specific Earth-Mars route.

                  Or, if one router had its time incorrectly set almost 30 minutes late, the other router would discard its packets as being twice as expired as above.

                  Or so I think.
          • by RockDoctor (15477)

            What is hard to understand here? There are no routers between Mars and Earth.. therefore the number of "modules" that process a datagram on the way from Mars to Earth is zero.

            That might be the current state of play. In the not too-distant future, the number of spacecraft in different parts of the solar system is going to require at the least several relaying stations in wide orbit. At the very least, two relays at the leading and trailing Trojan points would do away with the problem of regularly losing cont

    • That depends on how your routing protocols are configured.
  • by monopole (44023) on Monday March 03, 2008 @09:23PM (#22630616)
    ...Martian Packets [wikipedia.org]
  • ... to see if any computers that aren't on the net are thinking that we should do this. :)

    It sure sounds like the plot to Man Plus [wikipedia.org] might be coming true, excuse the bat wings.

    Art imitates life once again.
    • Art imitates life once again.
      Since the book already exists, shouldn't that be "Life imitates art once again"?
      • by rah1420 (234198)
        Of course it should. Can't decide what kind of affliction that is, so I'll just call it a 'screwup.'
  • That's it? One measly terabyte? For the whole planet?
  • How long until we get a terabyte solid-state dataserver (running IPv6, natch) in orbit around Mars?
    I don't know, but it'll be a bitch to drive there for each reboot.
  • How long until we get a terabyte solid-state dataserver (running IPv6, natch) in orbit around Mars?

    by suggesting IPv6 you've guaranteed it will never be implmented...

    • by tverbeek (457094)
      Now would also be a good time to get the RFCs in motion for establishing a TLD for Mars. The registration of helium.gov.mars is long overdue.

      And while we're at it, let's start referring to it as the System Wide Web, so we don't sound so old-fashioned and provincial.
  • Crossover point (Score:5, Insightful)

    by bradbury (33372) <Robert DOT Bradbury AT gmail DOT com> on Monday March 03, 2008 @09:33PM (#22630704) Homepage
    A better point for the exploration of the solar system is *when* can we set up a complete solar system that involves an information discovery and transmission system system which exceeds that which humans can carry out!

    The recent proposal to send humans to Mars is idiotic. I.e. we send take months and god knows how many $$ to send a few humans to Mars and then bring them back. What kind of an idiotic idea is that? One should be engaged (and I hope the folks at NASA are reading this) in a serious discussion of what is the information retrieval rate of a space probe (robotic explorer, etc.) vs.a human being?

    And so the discussion should be when the light speed transmission of information across the solar system will exceed the mass transport of humans across the solar system?
    • Re: (Score:3, Insightful)

      by MSZ (26307)
      One feature that humans have and (current) robots don't is the ability to understand situation (beyond simple rule-following) and have own decision. That is quite a good reason to send people on these missions.

      On the other hand, just sending a team there and getting them back after a week of exploration IS a waste of resources. It's just a kind of "my peni^Wrocket is bigger than yours"... It will end the same way as the Apollo missions to the Moon: we'll collect some data, plant a flag or two and then just
      • Re: (Score:3, Interesting)

        by pintpusher (854001)
        I'd like to see us send a significant number of cargo laden missions first. Sure stick a couple rovers on them to get some science done, but make the mission to get a critical mass of supplies up there first. Send a whole mess of them a month or two apart with the people in the last few. That gives you many launches to fix any problems that arise. You can make a reasonable effort to land them all within a certain radius. If you lose a couple on the way, it's okay because they're just supplies. YOu can alway
        • by khallow (566160)
          This is a good point, but I think any serious plan will have considerable preplaced supplies already on Mars. For example, the Mars Direct plans have a working nuclear reactor and the fuel (extracted from the Mars atmosphere and possibly on board liquid hydrogen) to return to Earth in position before humans show up (and the manned capsule would have it's own resources in case).
      • by Plugh (27537)
        I'm all for a lunar base and for a manned Mars expedition. I just don't want government to attempt to manage either one -- it'll be insanely overbudget and everyone will have to pay for it in taxes. Far better to let those of us who WANT to contribute to such efforts, make such contribution as we see fit, and with the organization we feel is most likely to do the job properly. Government just sucks. At everything.
        • Far better to let those of us who WANT to contribute to such efforts, make such contribution as we see fit, and with the organization we feel is most likely to do the job properly.

          Reality check: most people wouldn't care, and you wouldn't ever get enough funds together. Fund Mission to Mars or see Britney's snatch on TV? You do know what most people will choose, don't you?

          • by Plugh (27537)
            The point is whether you want to force people at gunpoint to pay for things they would not choose to, "because they don't know what's best for them". That line of thinking leads down a very dark path
      • by RobBebop (947356)

        There's no point in going to Mars unless we go there to set up forward base for colonization.

        I wrote about this the other day. See here [metaphrast.com].

        Here is the thought that sums up my meanderings...

        But with life on the Moon or Mars, I think we can't afford to assume that resupply missions will always be possible. I believe in providing the astronauts with a system to produce their own resources.

        I think that the hardest question for celestial colonization is how to design that system.

    • Re:Crossover point (Score:5, Interesting)

      by ceejayoz (567949) <cj@ceejayoz.com> on Monday March 03, 2008 @10:23PM (#22631060) Homepage Journal

      One should be engaged (and I hope the folks at NASA are reading this) in a serious discussion of what is the information retrieval rate of a space probe (robotic explorer, etc.) vs.a human being?
      http://astrobiology.arc.nasa.gov/news/expandnews.cfm?id=849 [nasa.gov]

      According to Kathy Clark, chief scientist for NASA's Human Exploration and Development of Space (HEDS), while the Sojourner Mars rover was a tremendous achievement, "Sojourner spent two weeks analyzing half a dozen Mars rocks. A human geologist could have done that same work in 30 minutes--then turned the rocks over to see what was hiding underneath."
      A biased source, but it's probably true - a human could travel the four miles Spirit has travelled in several years in about an hour. A little slower considering the sampling they'd be doing, but not by that much - you can pick it up and look at it on the way home (or when you get back to Earth). They could kick a deeper hole with their shoe in seconds than the rover can dig, ever.
      • That's a bit like saying I should buy a castle rather than live in my apartment, since it's a much nicer place to live. While true, that completely ignores price.

        Similarly, a human can do much more than Spirit, but you can probably send 10000 rovers like that for the same amount it costs to send one human.
        • Re: (Score:3, Interesting)

          by R3d M3rcury (871886)

          "Similarly, a human can do much more than Spirit, but you can probably send 10000 rovers like that for the same amount it costs to send one human."

          That's a good point--the question is, will you have fewer limitations?

          Human beings are pretty good tool users. So you send a geologist with lots of geology tools. He can wander across the plains of Mars looking for interesting rocks. Those rocks he finds that are interesting, he can bring back with him to the base where he has more tools. He can chip off a bit of rock with a hammer and look at it under a microscope.

          In short, figure out how to keep said geologist alive and productive on Mars and you've

          • by Gorimek (61128)
            1 or 100 robots can travel across the same plains with the same tools. They can't determine what is interesting, but can send data to people who can, and get instructions on what to do in 10-30 minutes.

            My point is that it's not "the same money". If it was, by all means send some adventurers. But it is actually several orders of magnitudes difference.

            I don't know about light/heavy rocket stuff. I'd go for whatever works best for least money and effort.
        • by khallow (566160)
          While there's economies of scale, you are just throwing out fantasy numbers. Keep in mind that if you can permanently settle a few hundred people on Mars, then they can grow to fill the planet. And there's only so much 10,000 rovers can do when they are controlled from Earth.
          • by Gorimek (61128)
            Yeah, 10,000 is a number out of thin air. Let's see... The internet says that the twin rovers cost $880M. I didn't quickly find a quote for a Mars expedition, and it would just be a wild guess now anyway, but the International Space Station is supposedly $130B. If going to Mars is an order of magnitude more, that's a factor of 3000.

            I wouldn't send 10,000 identical rovers to Mars. But you could do thousands of similarly complex robotic missions for the price of one human. Yes, humans are better in many ways.
            • by khallow (566160)
              In the past, NASA has claimed it could fund a Mars mission for $100 billion. Zubrin says he could do it for around $4-10 billion (I forget the exact figure). My take is that the price of a manned mission to Mars will go down considerably once some sort of economies of scale happen. Launching 10,000 rovers to Mars would do just that. I certainly wouldn't mind seeing that sort of effort even if it weren't directed at Mars. Everything gets cheaper as a result.
              • by Gorimek (61128)
                That's good to hear. I know the Google guys have talked about doing it. Either one could afford $4-10B.

                Sending rovers and other robots will no doubt also get much cheaper over time.
        • by RobBebop (947356)

          Similarly, a human can do much more than Spirit, but you can probably send 10000 rovers like that for the same amount it costs to send one human.

          In 30 years, there have been 130 Shuttle flights [wikipedia.org]. Two of these flights have been disasters. Granted, it is *easier* to send one rocket with a Rover on it per month... but saying "10,000" is a bit obtuse.

          What I would like to see is research aimed at sending robots that can build a habitable structure so the astronauts can survive for 6 months once they eventually make it to Mars.

      • More rocks.

        Unlike Discworld, on Mars, its rocks, all the way down. The ones studied by Soujourner are among the most expensive rocks in history. The ones studied in that 30 minutes would BE the most expensive rocks in history. But they'd still be rocks!
      • The era of human based space exploration has been and gone. Like all technology of a bygone era, it's remembered by many with fondness and sentimentality. Like all outdated technology, it has proponents who point to one thing or another that it does well, and then claim that modern technology is a step backwards. And now we have the car analogy. When cars first emerged, there were many who claimed that horse drawn transport was superior. They highlighted the advantages of horse drawn transport over cars. Fo
    • by Kjella (173770)
      We've got the probe thing down just fine, we've sent them as far as they can go interplanetary and interstellar is a completely different ballgame. I'd say the ability to shuttle people around the solar system (well, there's honestly not that many candidates but Mars) seems like a logical next step. There's a *lot* of long term settlement issues that have to be worked out for a Mars trip, that we'll never get started on unless we go. Waste of time and money? Well, not much more so than other basic research.
    • by khallow (566160)

      A better point for the exploration of the solar system is *when* can we set up a complete solar system that involves an information discovery and transmission system system which exceeds that which humans can carry out!

      Right now the humans are the decision makers and that is unlikely to change intentionally. Any setup that leaves humans on Earth and tries to do anything realtime is fundamentally flawed (once you get past a few light seconds and the orbit of the Moon) no matter how much AI you put in.

      Second, an unmanned only approach creates a high threshhold to doing anything in space. That is, you need to have considerable infrastructure and software in place to support a remote operated probe. Supporting a much larg

    • by tompaulco (629533)
      months and god knows how many $$ to send a few humans to Mars and then bring them back.
      I think I see an easy way to save half the time, and probably more than half the money.
    • by evanbd (210358)

      what is the information retrieval rate of a space probe (robotic explorer, etc.) vs.a human being?

      Currently, the *information* retrieval rate, potentially distinct from the *data* retrieval rate, is much, much higher for a person. The reason is that our ability to learn on Mars is limited by our tools, and robotic tools are far less capable than human-operated ones. This becomes especially true when you account for the fact that you can give a person instructions like "gather data on x, y, z, and anyt

    • The recent proposal to send humans to Mars is idiotic. I.e. we send take months and god knows how many $$ to send a few humans to Mars and then bring them back. What kind of an idiotic idea is that? One should be engaged (and I hope the folks at NASA are reading this) in a serious discussion of what is the information retrieval rate of a space probe (robotic explorer, etc.) vs.a human being?

      The discussion has long been engaged and has been over for almost as long - human explorers outperform robotic ones b

    • by autophile (640621)

      The recent proposal to send humans to Mars is idiotic. I.e. we send take months and god knows how many $$ to send a few humans to Mars and then bring them back. What kind of an idiotic idea is that?

      It's the kind of idiotic idea that will help us figure out what we need to do next time to send more humans to Mars.

      --Rob

  • I never get tired of that.

    I suppose the biggest question is, how long until CCP gets an EVE server up there?
  • by Provocateur (133110) on Monday March 03, 2008 @09:34PM (#22630712) Homepage
    PirateBay looks on with keen interest...
     
    • Re: (Score:3, Interesting)

      by Tastecicles (1153671)
      That actually raises a few questions of a legal, political, physical and commercial nature. Ergo;

      1. Who does the scientific data belong to? OK, most likely the hardware will be Chinese or American (built in Taiwan tho), and after some amount of political wrangling some of the Apollo/Ranger data was shared with other superpowers eventually. Who owns the raw data being stored or relayed by said hardware?

      2. As to commercial storage. What would the legal position be if some entity like TPB had a hosting platfor
      • by Bios_Hakr (68586)
        1. Scientific data belongs to the US taxpayer. Well, if NASA is doing the research, it does. There are times when the data won't be released. If the data is classified, for instance. In fact, NASA's budget is constantly replenished by the DoD. DoD pays NASA to send payloads and perform maintenance. For the most part, the data will be publicly available.

        2. If I buy a server in space, it will have to down-link somewhere. The **AA can just get a court order on the down-link provider. As for NASA/Gove
      • 3. The United States Government were mighty white to share the technology embedded in their GPS system for commercial use...

        I haven't heard that phrase used since the '80s. It seemed to me to have a dual meaning -- on the surface, a compliment, if you accept the underlying dynamic. But it's a backhanded compliment, implying that you've done something that is so obvious that you'd have to be pretty goofy *not* to do it. That's the disturbing part -- if you hadn't done it, you'd be "not white".

        The opposite
        • I was born in the 70's. I got a million words the whippersnappers of today don't know. :)
  • "4.294 billion ip's ought to be enough for any planet"
  • the plot of Doom 3?
  • hell yes! (Score:2, Funny)

    by ILuvRamen (1026668)
    I told my parents, when I move out I'm only going to live in a place that's got broadband. So I'm not moving to Mars until they got broadband! And none of that 8 minute ping time stuff. I can't play Starcraft like that. Run the backbone through a wormhole!
    • by calebt3 (1098475)
      Let's send a small herd of Slashdotters and South Korean gamers to live on Mars for the StarCraft finals. And let them stay. I'll be happy as long as I can get my /. and ISOs.
      • by tompaulco (629533)
        Let's send a small herd of Slashdotters and South Korean gamers to live on Mars for the StarCraft finals. And let them stay. I'll be happy as long as I can get my /. and ISOs.

        I don't think they'd notice the scenery change.
        "One town's very like another when your heads down over your pieces, brother."
        • Yeah, but it's a drag, it's a bore, it's really such a pity, to be looking at a board, not looking at the city... or marscape as the case may be.
  • ..somewhere with a worse ping time than us! (As long as the mars rovers play UT / WOW, I will actually be able to 0WN!!)
  • high-speed data transmission link from Phoenix to Earth during the landing
    Why do dump trucks and tubes come to mind?
  • Building an IT Infrastructure Around Mars sounds like an excellent new topic for "Ask Slashdot". There are enough experts here in the most random scientific arenas to discuss something hypothetically very cool.
  • It was in the works for a comm/gps system for Mars. Once W. invaded Iraq, he cut all budgets. This sat system went out the door. Now, we are looking at using the nano sats for doing something similar around the moon. It is possible that the same system will be replicated at mars. In particular, a number of techs were and are being developed to make this work. We now have a cheap, small, low energy nuclear clock. In addition, we are seeing efficiency increases in solar cells as well as energy storage system
  • There is already RFC supporting transmitting data over large-delay lines (e.g. from Mars). See RFC 1149.
  • This seems to fit in nicely with the MarsDirect idea, which revolves around having a lot of the infrastructure for a manned mission already in place before the astronauts leave Earth. I'd flat-out love to see us take the first real step off the planet before I croak.

    We'll have to do it sooner or later, anyway. Why not now?

    • I'd flat-out love to see us take the first real step off the planet before I croak...Why not now?
      Why? Did they find oil in Mars?
      Either that, or they should find a couple of terrorists in Mars for the Marines to invade.
  • We're sending up all those probes to build a network and prepare for the first manned exploration and (eventually) colonization.

    Has anyone realized that if Mars were inhabited, this would essentially be War of the Worlds in reverse?

    Little green people, prepare to be invaded!
  • The article could leave one with the impression that this is an 11th hour ad hoc improvisation, when nothing could be further from the truth. Using existing [Martian] satellites as relay stations for other satellites and landers has been part of the plans for at least a decade.
  • Take a look at http://tools.ietf.org/html/draft-irtf-ipnrg-arch-01 [ietf.org] and check out cool ASCII-art figures about example topology. [ietf.org] (Basically, they are thinking about adding a "bundle layer" between IP and TCP to make TCP work without timeouts becoming issue..)
  • Guess we'll have to accept martian packets [wikipedia.org] now.

It's a poor workman who blames his tools.

Working...