Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

A High-Bandwidth Interplanetary Connection

timothy posted about 3 years ago | from the tell-your-data-to-get-bent dept.

Communications 99

sciencehabit writes "A new study suggests that by twisting laser light, scientists could pack enough information into interplanetary beams to speed up extraterrestrial communications to the multi-gigabit level. The pulses would be passed through a hologram or multimode optical fiber, which twists the light. On the other side, a telescope would focus the light and a second hologram, or fiber would decode the signal. That could allow much more data-rich communication between, say, Earth and probes on Mars, the researchers say. Closer to home, the approach could provide Internet links of 100 gigabits per second."

cancel ×

99 comments

Size does matter... (3, Insightful)

RandomFactor (22447) | about 3 years ago | (#36784974)

> Hemmati and his colleagues estimate that receiving OAM data from a transmitter as distant as the sun would require a kilometer-wide telescope.

Sounds like even someplace closer like Mars is going to take an impractically large receiver.

Re:Size does matter... (2, Informative)

Anonymous Coward | about 3 years ago | (#36785054)

That's huge. Arecibo [wikipedia.org] is only 0.3 km.

Re:Size does matter... (1)

Anonymous Coward | about 3 years ago | (#36785080)

More than 50% of the time mars is not closer to earth than sun.

Re:Size does matter... (1)

Olix (812847) | about 3 years ago | (#36785240)

If you don't have to worry about gravity, making really big receivers and transmitters isn't much of an issue.

Re:Size does matter... (1)

lennier1 (264730) | about 3 years ago | (#36786512)

But at a size like stuff like statics is on a completely different level. Imagine the stresses on such a huge networked frame during something as unseemingly as a fine-adjustment of the array's rotation. Ass ume a decade of use and these things develop into a serious factor.

Re:Size does matter... (1)

Quietlife2k (612005) | about 3 years ago | (#36785740)

Is a relay via the moon not possible ?

We'd not have so much of a problem with size up there, and the link from here to the moon would require smaller telescopes, with at least one at each pole.

Re:Size does matter... (1)

HiThere (15173) | about 3 years ago | (#36787764)

I think an Lagrange point would be a better choice than the moon. The moon has too much gravity. But, as was pointed out earlier, rotating the antenna (telescope) would be a bit problem, even there. The lighter the structure, the less twist it could take.

Re:Size does matter... (1)

Boiling_point_ (443831) | about 3 years ago | (#36791312)

Sounds like even someplace closer like Mars is going to take an impractically large receiver.

The Sun is roughly the same distance from Earth all the time, because we have a roughly circular orbit around it.

Mars is sometimes closer to us when our orbital position around the Sun is on the same side as Mars, however due to the different year lengths on Mars and Earth, sometimes it's much further away on the far side of the Sun to us.

Re:Size does matter... (0)

Anonymous Coward | about 3 years ago | (#36791814)

The simple answer to this would be to put a satellite or two in a daisy chain between the two planets much like we do with mountain top receivers and transmitters. The laser could be received, error checked, and then passed on at full power once again, reducing the spread of the beam and improving power levels. This of course is using current technology which probably won't be around for much longer if trends continue.

You serious? (2)

jmd_akbar (1777312) | about 3 years ago | (#36784986)

We need faster speeds here down on earth before we think of these "multi gigabit" speeds for interplanetary communications..

Re:You serious? (3, Insightful)

John Hasler (414242) | about 3 years ago | (#36785018)

Why? How does thinking about "multi gigabit" speeds for interplanetary communications conflict with you getting a faster connection to Pirate Bay?

Re:You serious? (0)

Osgeld (1900440) | about 3 years ago | (#36786934)

cause they are not thinking about how to do it here, which is totally different than a line of site irda bullshit idea (which is fucking stupid in the first place, a laser will diffuse to be larger than the planet you want to aim for)

its a simple matter of quit wasting our time and money for these mankind space cloud ideas and lets fucking finish something down here

Re:You serious? (0)

Anonymous Coward | about 3 years ago | (#36788756)

I'm sure you are just a troll, but lots of communications systems here on earth use Line of Sight. Also how do you know they are not thinking about how to do it here? How many scientists and engineers are we supposed to dedicate to a single subject? If there are only say 10,000 people working on improving local communications is that still a problem for you? And what about the many potential improvements in fiber optic we've been seeing on Slashdot over the last yeasr? Do they not count?

Re:You serious? (1)

slick7 (1703596) | about 3 years ago | (#36788278)

Why? How does thinking about "multi gigabit" speeds for interplanetary communications conflict with you getting a faster connection to Pirate Bay?

(sarcasm) Since the government refuses to confirm or deny the existence of alien life forms, this is just a waste of money that should be going to the war effort. (/sarcasm)

Re:You serious? (1)

RoFLKOPTr (1294290) | about 3 years ago | (#36785406)

We need faster speeds here down on earth before we think of these "multi gigabit" speeds for interplanetary communications..

We HAVE faster speeds here on Earth. Blame your ISP for not utilizing what their network is capable of, preferring to intentionally cripple their services so they can charge through the nose for faster connections. But all that is really totally irrelevant because this Earth-Mars link is not in any way intended for consumer use.

Re:You serious? (1)

Idbar (1034346) | about 3 years ago | (#36785532)

I'm crossing my fingers. Hopefully, I'd be able to get 100G first from a martian ISP than the crappy US ISPs. And furthermore, I hope is not going to be capped by the ISP and government regulations or tapped **IA.

Re:You serious? (1)

hlavac (914630) | about 3 years ago | (#36786602)

If you don't mind the ping between 6 and 44 minutes and the ocassional outage due to Sun getting in the way, then by all means ;)

Re:You serious? (0)

Anonymous Coward | about 3 years ago | (#36795454)

100G DWDM is already running on backbone links with up to 8.8Tbps per physical fiber pair. Don't believe me, see this Ciena marketing video for the US Internet2:
http://www.youtube.com/watch?v=hYd8pjf3z4A

Re:You serious? (1)

Lord Lode (1290856) | about 3 years ago | (#36785572)

If they send people to Mars, they better still be able to play in multiplayer with their friends on Earth! Not to mention Youtube.

Re:You serious? (0)

Anonymous Coward | about 3 years ago | (#36786292)

The delay would be way to long, genius.

Re:You serious? (1)

mr_mischief (456295) | about 3 years ago | (#36786546)

Of course you're joking, but it's not an all-around impossible idea.

Bandwidth? Check. Ping? Not so much. Or rather, too much. Or something. The bandwidth being high doesn't lower the latency any of course.

"The communications delay between Earth and Mars can vary between five and twenty minutes depending upon the relative positions of the two planets. As a consequence of this, if a robot on the surface of Mars were to encounter a problem, its human controllers would not be aware of it until at least five minutes, and possibly up to twenty minutes, later; it would then take a further five to twenty minutes for instructions to travel from Earth to Mars." from this Wikipedia article on the speed of light [wikipedia.org] might be useful for someone.

So chess between Mars and Earth? Sure. Fantasy football? Sure. Half-life or Supreme Commander? Not so much.

YouTube? Well, queue up a bunch of them to stream and start streaming each one five to twenty minutes before it needs to start playing. The initial wait will be a bit inconvenient, but there's no reason with that much bandwidth you couldn't stream multiple videos simultaneously while waiting.

Re:You serious? (1)

paulo.casanova (2222146) | about 3 years ago | (#36791496)

I would imagine if we had serious computing business in Mars we would have some sort of caching running there :) Like, full replication... :) Just an idea, of course...

Re:You serious? (1)

mr_mischief (456295) | about 3 years ago | (#36793072)

Sending people to Mars does not imply large-scale long-term colonization.Serious business wouldn't be built to keep a handful of researchers connected to the homeworld. If there was sufficient demand for content, though, sure you could expect both caching much of Earth's content on Mars as well as a large cache on Earth of content from Mars.

Are they economically justified? (0)

Anonymous Coward | about 3 years ago | (#36786594)

Most residential customers are cheap, and satisfied with a few megabits per second for their web browsing, and watching Youtube videos. Would they pay for much more speed than they currently need?

As for Mars, a multigigabit connection could allow for lots of Mars pictures for scientific studies. Telepresence would be very helpful on Mars. Put the money down for Mars.

Re:You serious? (1)

petermgreen (876956) | about 3 years ago | (#36787836)

Afaict we have 10 gigabit (maybe more now) through a single optical transciver and we can combine multiple such links on a single fiber through WDM and a typical cable contains many fibers.

The big issue with end user connections is not technical it's financial, upgrading all the connections to end users is very expensive and most users probablly aren't prepared to pay all that much more for a faster connection. So the incentive for providers to upgrade is low. Further any sensible last mile communications provider knows that the longer they hold off upgrading the better the technology they upgrade to will be.

Ping times (3, Informative)

s_p_oneil (795792) | about 3 years ago | (#36785008)

They're going to have to do something about the terrible ping times. Its orbit is about 1.5AU, so when it's close to the Earth, the round-trip ping time will be about 8 minutes. When it's on the opposite side of the sun, it'll be about 40 minutes.

Re:Ping times (1)

Anonymous Coward | about 3 years ago | (#36785228)

After they're done with this, I'm sure they'll start working on getting around that pesky ole' theory of relativity...

Re:Ping times (0)

Anonymous Coward | about 3 years ago | (#36785246)

Short of changing the laws of physics, there is nothing to fix those ping times.

Re:Ping times (1)

SuricouRaven (1897204) | about 3 years ago | (#36785354)

Bit more than that - the sun would blind any attempts at communication when mars is directly opposite. You're going to have to either settle for being out of contact for a short time, or bounce the signal from somewhere else. The other inner planets arn't very suited to building a communication station, so probably a router at earth-sun L4/5 point.

Eight minutes means we'd have to dump this 'stream everything' internet and start actually downloading files before watching them again.

Re:Ping times (1)

h4rr4r (612664) | about 3 years ago | (#36785804)

No it does not. Netflix-Mars would just copy all their data from Netflix-Earth once a $time-interval and the Martians would just stream from Netflix-Mars.

Re:Ping times (1)

TheLink (130905) | about 3 years ago | (#36786048)

But you're proposing planet scale copying! Copying is illegal!

The Corporations unwritten constitutional right to profits must be protected!

Re:Ping times (1)

hedwards (940851) | about 3 years ago | (#36786160)

Actually, it's easier than that, just send a DVD or two to mars ever year and they'd be good to go. Plus, they can raise rates periodically because you ought to be grateful that they're letting you have all but season 3 of the Dukes of Hazard by stream.

Re:Ping times (4, Funny)

pushing-robot (1037830) | about 3 years ago | (#36785748)

Yeah, the latency on wireless connections always sucks. If we plan on being on Mars a lot, we should just go ahead and string a cable.

Re:Ping times (1)

rubycodez (864176) | about 3 years ago | (#36786940)

the spins of earth and mars should present no problem just as long as we remember to unwind the cable periodically, I see no big obstacles to your idea. We can save money by only unwinding the Mars end and giving the cable a big up and down shake to undue the earth end.

Re:Ping times (1)

Nethemas the Great (909900) | about 3 years ago | (#36800674)

Except for the glowing ball that sometimes stands between the two...

Re:Ping times (0)

Anonymous Coward | about 3 years ago | (#36791466)

Latency depends here mainly on distance, not on wired / wireless. A cable to Mars, if somehow made possible, would have higher latency then laser communication, as the speed of electromagentic waves is lower in a cable than it is in vacuum.

Re:Ping times (1)

Rene S. Hollan (1943) | about 3 years ago | (#36785802)

I was going to say, "Yeah but latency would S.....U.....C.....K."

Re:Ping times (1)

Quietlife2k (612005) | about 3 years ago | (#36785948)

I might be wrong, but surely it's a 4 SECOND ping time ?

Earth to Mars furthest distance 401 million km.

Speed of light 299 million km per second.

Round trip time 4 SECONDS.

Ok so at this distance the sun is smack bang in the way, but from all other viable viewing points the distance is less than the 401 million km used above.

Re:Ping times (1)

cranq (61540) | about 3 years ago | (#36786066)

Speed of light is 299,792,458 metres per second, not km per second.

Re:Ping times (2)

Dwonis (52652) | about 3 years ago | (#36786078)

Speed of light is 299,792,458 metres per second, not km per second.

Sure, but by the time these machines get built, Moore's Law will have sped it up by the necessary amount.

Re:Ping times (1)

Quietlife2k (612005) | about 3 years ago | (#36786082)

Aww shucks got my decimals in the wrong place :-(

You are correct my apologies.

Re:Ping times (1)

TheLink (130905) | about 3 years ago | (#36786106)

No. Speed of light is only about 300 million metres per second ( 299 792 458) or 300000km/sec.

http://www.google.com/search?q=speed+of+light [google.com]

So round trip time is 44.6 minutes
http://www.google.com/search?&q=401+million+km++*+2+%2F+speed+of+light [google.com]

In comparison the Sun is about 8 light minutes away.

The speed of light is so slow that even latency is an issue for intercontinental undersea links, and worse for the satellite links (which can have latency in the order of seconds).

Re:Ping times (1)

TheLink (130905) | about 3 years ago | (#36786128)

BTW yes I know geostationary satellites are 40000km away and the theoretical round trip ping latency via pure satellite links is thus 280ms * 2 . But in practice it sure seems worse than that ;).

Re:Ping times (1)

Aqualung812 (959532) | about 3 years ago | (#36787198)

I'm surprised no one has discussed TCP window size. The TCP stack on both ends will need a huge amount of RAM, and even with that, I'm not sure how they are actually going to get the throughput needed if we're talking TCP/IP.

Perhaps UDP or some non-IP method. Remember, bandwidth != throughput.

Re:Ping times (1)

petermgreen (876956) | about 3 years ago | (#36787950)

Any high speed interplanetary connection is going to either have to put up with occasional data loss (how occasional can be influenced to some extent by forward error correction but that comes at a price in bandwidth and still doesn't really provide a soloution for block outages) or have lots of storage for retransmissions.

TCP is not appropriate, it was designed to deal with connections of relatively low latency and unknown bandwidth by slow-starting. On a high latency connection of known bandwidth that is just plain stupid. Also with such high ping times keeping the retransmit data in a ram buffer rather than on cheaper disk or flash is also crazy.

Re:Ping times (1)

Hamsterdan (815291) | about 3 years ago | (#36787424)

Seems like DSL with Bell.

Re:Ping times (1)

Dadoo (899435) | about 3 years ago | (#36788684)

They're going to have to do something about the terrible ping times.

I'm guessing IP is out, since the latency will be horrendous. They'll probably have to bring back a mechanism like UUCP, where your files get dropped off in a temporary storage server. There, they'll be placed in a queue to be transmitted to another planet, and you'll be emailed, once transmission is complete.

Mars ain't the kind of place to raise your kids (0)

rossdee (243626) | about 3 years ago | (#36785014)

The farthest away people are is in LEO (the ISS) So why do we need to develop faster links to other planets.

Re:Mars ain't the kind of place to raise your kids (0)

Anonymous Coward | about 3 years ago | (#36785044)

But there is equipment that is further away and some of that equipment still has restricted transfer windows during which to transmit/receive data.

Re:Mars ain't the kind of place to raise your kids (1)

seantide (2378542) | about 3 years ago | (#36785058)

Article: That could allow much more data-rich communication between, say, Earth and probes on Mars, the researchers say.

Re:Mars ain't the kind of place to raise your kids (2)

mbone (558574) | about 3 years ago | (#36785668)

Because most space probes now-a-days are data rate limited. LRO, MRO, Dawn, etc., all could take more data, if we could get it back.

Re:Mars ain't the kind of place to raise your kids (0)

Anonymous Coward | about 3 years ago | (#36786020)

In fact, it's cold as hell. And there's no one there to raise them, if you did.

Is it practical? (1)

craigminah (1885846) | about 3 years ago | (#36785052)

While this seems simple enough, maintaining a laser oriented at Mars while each planet rotates on its own axis and around the sun would made this impractical unless there were multiple transmitters/receivers at each location/node along with a terrestrial network to interconnect them.

Re:Is it practical? (0)

Anonymous Coward | about 3 years ago | (#36786184)

We can shoot straight through the sun: just use an x-ray laser.

Cat5 (0)

Anonymous Coward | about 3 years ago | (#36785076)

As seen on slashdot good old Ethernet is still alive. Cat5 would do thejob, but you need to give it some slag to allow for distance variations.

Optical is taking a long time... (1)

mbone (558574) | about 3 years ago | (#36785094)

There is as yet no interplanetary communication by Laser. It's all done by radio at present. The first flight demonstration [mit.edu] of Laser communication will be on the LADEE [wikipedia.org] Lunar Orbiter. That's scheduled to be launched in 2012. I am sure that optical communications will eventually be used, though. Using reasonably sized telescopes, gigabit per second communication across interplanetary distances should be possible using conventional techniques, even if OAM is not actually practical. (Of course, the weather would be an issue, as laser signals can't go through clouds.)

Given that more or less the same thing could have been and was said about deep space optical communications in the mid-1980's, all I can say is that it sure is taking a long time to implement it.

Re:Optical is taking a long time... (1)

SuricouRaven (1897204) | about 3 years ago | (#36785364)

Is there that much demand for higher bandwidth, outside of earth-sats? The only machines further out than that are scientific probes, and all they need to send back is telemetry and the occasional photograph.

Re:Optical is taking a long time... (1)

mbone (558574) | about 3 years ago | (#36785686)

As I posted above, most scientific probes now are data-rate limited. They could acquire more data, but they can't send it back, due to data-rate and Earth-side antenna-availability limitations. So, yes, there is a strong demand for higher bandwidth.

how will we talk to the aliens? (1)

decora (1710862) | about 3 years ago | (#36786488)

if they are trying to send us a new video of their greatest pop star, how will we be able to download it in a reasonable amount of time?

Re:Optical is taking a long time... (1)

imsabbel (611519) | about 3 years ago | (#36786966)

Well... if they had more bandwith, they could send more than the occasional photo.

Those landers on titan only could send 10Mbyte or so during the decent.

New horizons has a multi-GB solid state drive. It will fill it up during Pluto flyby (limiting the amount of data it can aquire), and will then spend _months_ transmitting it back a modem speeds.

With a real time data link, those probes could have / would be able to transfer orders of magnitude more data back home.

Throttling (4, Funny)

Rik Sweeney (471717) | about 3 years ago | (#36785112)

Closer to home, the approach could provide Internet links of 100 gigabits per second

Throttled down by your ISP to 24 megabits per second

Re:Throttling (3, Insightful)

nine-times (778537) | about 3 years ago | (#36785146)

... with a 768kbps upload rate.

Re:Throttling (0)

Anonymous Coward | about 3 years ago | (#36785152)

And a cap of 40GB

Re:Throttling (0)

Anonymous Coward | about 3 years ago | (#36785548)

..and deep packet inspection to monitor your connection...

Re:Throttling (0)

Anonymous Coward | about 3 years ago | (#36786696)

And a DMCA takedown order

Re:Throttling (0)

Anonymous Coward | about 3 years ago | (#36785296)

.....and we will charge you your home to get on the plan to pay for our dumb ass marketing staring Shaquille O'Neal then charge you additional increases to include your first born child, your pension, and we will try our best as your provider to squeeze every last cent of any other forms of undeserved (by us) revenue you make on your following statements and not even tell you about it. We are also planning with the increased revenue to lobby government to give us fatter grants then we have had in the past and as in the past continue to plan to show nothing for it and for legislation to allow us to tap into your social security to pay for future billing statements.

Re:Throttling (1)

Warwick Allison (209388) | about 3 years ago | (#36792036)

Only after you go over your 1TB cap, which would take... oh... sorry, too late.

At least... (1)

bytethese (1372715) | about 3 years ago | (#36785142)

They'll be able to stay occupied downloading music and porn quicker.

A ship full of hard disks? (2)

craznar (710808) | about 3 years ago | (#36785188)

Should give limitless bandwidth.

Latency of course is the factor here.

Re:A ship full of hard disks? (0)

Anonymous Coward | about 3 years ago | (#36785408)

Packet loss.

Re:A ship full of hard disks? (1)

xkuehn (2202854) | about 3 years ago | (#36786578)

Too true. The Martian IPBM system would wreak havoc with this.

Re:A ship full of hard disks? (0)

Anonymous Coward | about 3 years ago | (#36785496)

Let's see, a data pipe just 4 Light Years long filled with data packed to 100 Gb/s -- or 100 Gb in an 300,000 KM-long portion of that pipe... It'll still take 4 years to get there.

Re:A ship full of hard disks? (0)

Anonymous Coward | about 3 years ago | (#36785502)

According to this [howstuffworks.com] , it currently costs about $1000000 per pound to send something to Mars (I'll be conservative and say $400/g). Picking the first hard drive that Google gave me [pcmag.com] , we have 16Tb of storage at 655g, or 4.1e-11 g/bit or 1.64e-8 $/bit, about $164 per gigabit.

Setting up lasers and shit has a lot of upfront cost, but once you get it set up, I'm assuming it's considerably cheaper than $164 per gigabit in the long run.

Re:A ship full of hard disks? (0)

Anonymous Coward | about 3 years ago | (#36785564)

Replying to myself. That should be $16.40 per gigabit, not $164 per gigabit. Still kind of expensive, though.

Re:A ship full of hard disks? (1)

ffejie (779512) | about 3 years ago | (#36785918)

But, in twenty years, when someone actually needs this stuff on Mars, you'll be looking at $0.10 GB to get it up there, even assuming NASA or the like has to pay for a special "ruggedized" harddrive that is guaranteed to work in a different atmosphere. Storage costs will drop.

I think the real problem is getting it back. Right now, probes have a really slow data connection to send back data. Assuming we can blast off tons of harddrives to them doesn't help that side of the equation.

Also, I was going to make a witty comment about latency, but a few others beat me to the punch (no gaming on Mars), using "sneakerNET" to shuttle things back and forth gives possible the worst latency of all: years.

Re:A ship full of hard disks? (2)

Hamsterdan (815291) | about 3 years ago | (#36787458)

Only problem is by the time it gets there, the hard drive will have bad sectors in true Seagate fashion

That's great! (2, Funny)

Anonymous Coward | about 3 years ago | (#36785198)

That's great, but the roaming charges will kill you.

Re:That's great! (1)

hedwards (940851) | about 3 years ago | (#36786172)

I thought the bigger problem was that people were headed to Mars to avoid Arnold...

Ping (0)

JulianDraak (1918564) | about 3 years ago | (#36785212)

Ping times would be horrible in a TF2 game of Mars vs Earth.

HIgh bandwidth is easy... (1)

mark-t (151149) | about 3 years ago | (#36785236)

It's the latency that's the real killer.

If you want to just talk bandwidth... which is how much data can be sent from one location to another within a unit of time, then the bandwidth of a cargo van hauling a truckload of flash ram sticks across the country has, by many orders of magnitude, a higher bandwidth than any sort of technology that is built on electronic communication infrastructure.

When it becomes possible to, from earth, ping an orbiter around mars in under a second... *THEN* we will have really done something (okay... we may have violated a few physical laws to accomplish it, but hey... we can't let _that- stand in the way of technological progress, right?)

Re:HIgh bandwidth is easy... (2)

ffejie (779512) | about 3 years ago | (#36786166)

I love the dumptruck analogy, but I'm not convinced of the math. Let's see if we can work it out:

A dumptruck has a volume of approximately 722 cubic feet (17 x 8.5 x 5) source [yahoo.com] . Converting that gets us 1,247,616 cubic inches.

A harddrive is 3.5" x 102 mm x 25.4 mm source [wikipedia.org] . Or about 14 cubic inches.

This means that roughly, we can fit about 89,115 hard drives into a dumptruck, assuming everything fits perfectly.

The largest commercially available 3.5" harddrive is 3 TB. This means that we're going to have 267,343 TB on our truck.

Driving across the country takes 26 hours assuming no stopping. Source [google.com]

This yields 267,343 TB / 26 hours = 2,138,744 Tb / 26 hours = 2,138,744 Tb / 93,600 seconds = 22.849 Tbps = 22,849 Gbps.

Compare that to a commercially available 10Gbps link available from any business class provider, and you're going to trounce them. The latest stuff is 100Gbps, which you should be able to get a hold of if you're willing to shell out right now, but it's still a blow out. You are indeed correct.

A couple of other items to note:
1. You can add a day at the front and a day at the back to load and unload the truck and you're still around 10 Tbps.
2. If you add another 2 days to collect all the data, write it to harddrives and then do the same 2 day process at the back end, you're still around 5 Tbps.
3. A more fair comparison is probably not using such a slow mechanism like a dump truck (I understand it's there to prove a point in your analogy). But instead use a cargo plan. You'll get 23,200 cubic feet of storage out of a Lockheed C-5 Galaxy cargo plane [wikipedia.org] which is about 32 dumptrucks. Also, you can get across the country in 6 hours, even assuming some landing and takeoff. If you still assume it takes several days to load and unload the plane, you'd probably be up in the 1 Pbps range.
4. I was trying to think what would be the fastest way to load and unload the truck, and then I realized I was insane for trying to calculate such a dumb thing. The real answer here is to build storage arrays on the truck and have it roll back and forth. Essentially, it's a mobile data center. You have a few hundred 10 Gbps ports on the rear of the truck to plug in, and you can download all of your data on and off of it pretty quickly, without manual labor beyond plugging in a port. Further, this would be a very fun way for the Pirate Bay, or someone similar to distribute data if they ever wanted to go physical (and thus blatantly break the law). Once a week, the Pirate Bay truck would roll into town and all the kids could plug in the back to download petabytes of information before it leaves for the week. A fun concept, if nothing else.

Re:HIgh bandwidth is easy... (1)

xkuehn (2202854) | about 3 years ago | (#36786914)

Assumption fail. Volume is not necessarily a limiting constraint. Hard drives are dense.

According to this http://wiki.answers.com/Q/How_much_gravel_can_a_dump_truck_hold [answers.com] a dump truck can typically hold 13T. I've googled the weight of a 3.5" hard drive and I get values between 400g and over 1kg. So I weighed an old one, and got a bit over 400g. The 400g drives are those with fewer platters, but I'll go with it for simplicity.

13,000kg / 0.4kg gives you 32,500 hard drives. Continuing with your calculations, 3TB*32,500 = 97,500TB = 780,000Tb. (According to 'man units', hard drive manufacturers use SI units.) 780,000Tb/26h = 780,000Tb/93,600s = 8.333Tb/s or 8333Gb/s.

You'll get 23,200 cubic feet of storage out of a Lockheed C-5 Galaxy cargo plane which is about 32 dumptrucks.

23,200*(12^3)/14 * 0.4kg ~= 1,145,416kg The plane can hold a maximum of 81,600 kg. And fuel economy is horrible.

Not that I disagree with your general point.

Re:HIgh bandwidth is easy... (1)

PlazMan (40335) | about 3 years ago | (#36788230)

Driving across the country takes 26 hours assuming no stopping. Source [google.com]

I think you meant 50 hours.

DNA porn (0)

eggstasy (458692) | about 3 years ago | (#36785302)

100 gigabit connections will open up the vast unexplored market for molecular biology porn!
On a 15 inch monitor, you'd be able to fit enough pixels to make out ribosomes and cell walls, not to mention viruses and bacteria.
Pervs all over the world would devote entire discussion forums to their favorite STDs in porn stars.

Re:DNA porn (1)

rubycodez (864176) | about 3 years ago | (#36787002)

of course a slashdot reader would think anything of the "molecular biology" type of sexual entertainment would involve high bandwidth and looking at a monitor. I could think of another way that involves another person.....

Why bother... (1)

Anonymous Coward | about 3 years ago | (#36785350)

They'll just cap it anyway. :( I wish i was joking.

Speed or power? (2)

GooberToo (74388) | about 3 years ago | (#36785426)

Isn't the limiting factor for current bandwidth power rather than speed anyways? My understanding they've long had the technology for much high bandwidth but the limitations are always power demands.

Secondary to power is usability. There's a big difference between pointing an antenna in the general direction of home and precisely aiming a laser million of miles away. Several orders of magnitude more accuracy is required.

Personally I don't find anything practical about this project. At least not today.

Re:Speed or power? (1)

Migraineman (632203) | about 3 years ago | (#36785746)

[INT. ENTERPRISE BRIDGE]

Picard: Mr. LaForge, we're having trouble receiving the signal from the Very Far Away Observatory. Can you boost the signal.
LaForge (v.o.): Sir, we're already using a holographic multi-modal optical receiver.We're operating near the theoretical limit.
Picard: Prehaps you could route secondary power through the replicators in the galley on deck 12.
LaForge (v.o.): Uhm ... yeah ... I'll get right on that. LaForge out.


Power v. Bandwidth is always one of the spacecraft design tradeoffs. For signals with really low power at the receiver, smaller bandwidth can be beneficial because you can use filters (physical and DSP) to improve your SNR and your chances of recovering your data. Generally, you can't evaluate each piece in the comm link individually. The trick is to find a solution that meets your power budget, link budget, bandwidth objective, mass budget, volume budget ... all simultaneously.

If you want some interesting docs on space communications, check out the pubs from the CDSDS. [ccsds.org]

Re:Speed or power? (0)

Anonymous Coward | about 3 years ago | (#36789946)

Power is a limiting factor, but not in the sense of being a bottleneck. The technique being described allows them to getter higher bandwidth for the same amount of power.

You're right, the required pointing accuracy may be a problem. If you're using 5cm-wavelength radio with a 5 metre antenna, your probe's transmission beam is (5cm/5m) = 0.01 radians ~= 0.5 degrees across, which gives you a bit of slack. I don't know what it would be for optical communications, but I'd expect it to be narrower.

Re:Speed or power? (0)

Anonymous Coward | about 3 years ago | (#36795888)

The technique being described allows them to getter higher bandwidth for the same amount of power.

No it's not. Current high bandwidth laser comms using PPM (Pulse Position Modulation), so if you're using 8 position/3 bit encoding your beam is on for less than 1/8th of the time. OAM is a variant on QAM (Quadrature Amplitude Modulation) used to get us from 300bps modems to 600bps and higher - the beam will be on all of the time, just changing the amplitude and twisting its polarization. This will consume a lot more power than PPM equivalents.

yea go for it (1)

Osgeld (1900440) | about 3 years ago | (#36786712)

Interplanetary IRDA is just as dumb and useless as it is here for the exact same reasons

Better have a squid proxy (0)

Anonymous Coward | about 3 years ago | (#36787156)

And have your server be down loadable to remote nodes so you can accept input from users on mars. Maybe design it so that your cloud server can have a few nodes on mars and out in the asteroids. Then stream updates back and forth between cloud nodes using interplanetary IP.

Bummer (1)

vvpt (1077009) | about 3 years ago | (#36787326)

Sounds like inter-planetary multi-player Quake is still out of reach.

Interstellar Bandwidth Auction (2)

retroworks (652802) | about 3 years ago | (#36788010)

Rather than actually doing it, couldn't we just auction the interstellar bandwidth to Google, Verizon, ATT, Sprint, etc.? Then we'd close the federal deficit! Centuries from now, they'd make the money back on roaming charges. We've been passing the buck to our kids, time to exploit the great-great-great-grandkids.

It's not the bandwidth it's the latency (1)

davidwr (791652) | about 3 years ago | (#36789532)

OK, maybe bandwidth is an issue but latency is still a bear.

IPN Gets A New Layer 1 (0)

cmholm (69081) | about 3 years ago | (#36789878)

Obviously, this doesn't "solve" the latency issue, but the concept does help bandwidth. Also, it doesn't replace RF links, but merely would relegate them to the failover for the IPN [wikipedia.org] version of the BGP.

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...