Moon-Based Supercomputer Proposed

The old chums would have titled this paper: “Let’s Put a Dinkum Thinkum on Luna.”

Earth’s Deep Space Network of giant antennas used to gather data and talk to spacecraft is already overloaded, and the demand for bandwidth will only go up. Ouliang Chang, a grad student from my alma mater, USC, recently presented his solution at a space conference (, “Why We Need a Supercomputer on the Moon.”) —

The plan is to bury a massive machine in a deep dark crater, on the side of the moon that’s facing away from Earth and all of its electromagnetic chatter. Nuclear-powered, it would process data for space missions and slingshot Earth’s Deep Space Network into a brand new moon-centric era.

Factoring in the cost of launching components into space, he estimates the project would cost between $10-$20 billion.

That’s before we start hiring the specialists. Keep an eye open for this classified ad:

Wanted: one-armed computer tech to spend his time explaining which jokes are funny…

[Thanks to David Klaus for the story.]

3 thoughts on “Moon-Based Supercomputer Proposed

  1. Oh yes? No unforeseen problems or cost over-runs here… Other than it may be 10 or 20 years premature to suggest it, it’s not a bad idea.

  2. The problem seems to be the issue of how we’ll communicate with this thing in its deep dark crater on the the moon. The whole point of putting it there is, apparently, to shield it from Earth-originating transmissions, so not only would you need the stuff on the moon, but you’d also need a relay station with visibility of both the earth and the moon site. Given the lack of an equivalent to a geosynchronous orbit (due to the absence of lunar rotation), that implies a constellation of relay satellites in lunar orbit and a bunch of remote antennae on the moon to communicate with them.

    The idea is interesting, but it seems blinded by terrestrial thinking. Shielding an antenna doesn’t require a huge lump of rock if you have suitably large sheet of the appropriate material. So why not take this same supercomputer with its same graphene sheets and stuff it into a nice Lagrange point. Then unfurl a large sheet and and hide from whatever large radio sources you want to mask (the earth, possibly, but rather more likely that big yellow transmitter in the sky). Communicating back to earth just needs an antenna on the other side.
    Plus, rather than mess around with nuclear reactors, why not use the shield as a site for photovoltaic panels…

  3. Satillite links in lunar orbit is too fancy a solution. Cheap, line-of-sight microwave relays, from peak to peak would do the job splendidly. Of course, you would need some way of planting them there. Maybe satillites would be cheaper after all.

    Would it be necessary for the communications network be located at a Lagrangian Point? It isn’t as though it would tax even your cell phone to calculate where in orbit, to aim a signal at a given time.

Comments are closed.