Lumen Orbit - Data Centers in Space

[Michel Van]

The company Lumen Orbit want to put Data servers,
to use the abundant solar energy, cooling, and the ability to freely scale up in space.
launch of Prototype hardware is set for may 2025 launch of first micro data center in 2026
once Starship goes into commercial service they build the 5GW Data server in Low orbit

View: https://www.youtube.com/watch?v=-T--N58vwoo
Note: the original Video by Lumen Orbit is removed from YouTube...

Source:

Launch YC: Lumen Orbit ð Data Centers in Space | Y Combinator

Lumen Orbit is building data centers in space to make use of 24/7 solar energy and passive cooling.

www.ycombinator.com

 

[aim9xray]

I suppose that would rule out visiting your CoLo server to reboot?

 

[TomS]

The company Lumen Orbit want to put Data servers,
to use the abundant solar energy, cooling, and the ability to freely scale up in space.
launch of Prototype hardware is set for may 2025 launch of first micro data center in 2026
once Starship goes into commercial service they build the 5GW Data server in Low orbit

View: https://www.youtube.com/watch?v=-T--N58vwoo


Source:

Launch YC: Lumen Orbit ð Data Centers in Space | Y Combinator

Lumen Orbit is building data centers in space to make use of 24/7 solar energy and passive cooling.

www.ycombinator.com

Abundant cooling? In space? Are they aware that this is functionally like building a server in a thermos bottle?

 

[Michel Van]

Abundant cooling? In space? Are they aware that this is functionally like building a server in a thermos bottle?

i'm afraid not, there Data Center concept art has no radiators...

 

[Justo Miranda]

In my opinion this thing does not seem very safe to preserve or exchange data, regardless of temperature changes it would need a very heavy radiation shielding... What is the advantage?

 

[Grey Havoc]

Not to mention little things like ASATS....

 

[TheKutKu]

Not-A-Starship! With delta flaps...
Of course, you don't want the solar powered data center to be in such a low orbit.

 

[TomS]

View attachment 739787

Not-A-Starship! With delta flaps...
Of course, you don't want the solar powered data center to be in such a low orbit.

You do if you're trying to manage latency, which is kind of a big deal for servers.

This is such a dumb idea.

 

[Dilandu]

What is the advantage?

They can't be physically seized by police, and their destruction would cause too many international outcry to seriously consider it as police measure.

 

[Dilandu]

You do if you're trying to manage latency, which is kind of a big deal for servers.

This is such a dumb idea.

Well, it could use polar orbit parallel to terminator, to stay in sunlight as much as possible.

 

[Yellow Palace]

They can't be physically seized by police, and their destruction would cause too many international outcry to seriously consider it as police measure.

Also, the locals can't complain about the size/noise/power requirements for the data centre. Which is also becoming an issue. Though even for a 'small' data centre you're talking about megawatts of power, and one in space would need even more to run the cooling.

To the extent that this is an idea, rather than merely a press release, I suspect it's a reaction to it being increasingly difficult to find sites that are suitable for, and willing to accept, terrestrial data centres. Microsoft was doing some work on subsea data centres for similar reasons, which seemed pretty promising.

 

[TheKutKu]

They can't be physically seized by police, and their destruction would cause too many international outcry to seriously consider it as police measure.

On the other hand, the destruction can be done by any motivated and skilled decently sized team no matter the country.

 

[TomS]

Not to mention little things like ASATS....

On the other hand, the destruction can be done by any motivated and skilled decently sized team no matter the country.

ASAT is hardly a trivial undertaking and can't be done by a small team, no matter how motivated or skilled. It takes a moderately large pile of money, the kind of money associated with nation-state actors.

There are four countries with demonstrated ASAT capacity, and I can think of maybe four more that could do it in relatively short order, as a national effort.

 

[TheKutKu]

ASAT is hardly a trivial undertaking and can't be done by a small team, no matter how motivated or skilled. It takes a moderately large pile of money, the kind of money associated with nation-state actors.

There are four countries with demonstrated ASAT capacity, and I can think of maybe four more that could do it in relatively short order, as a national effort.

ASAT and Targeting a thin, City sized LEO target with no maneuvering capability are two completely different thing.

Basically any decently sized sounding rocket does the job of significantly damaging the array and particularly polluting its orbit.

So this needs active space based defense, laser won't cut it at diverting the mass away from the kilometer-sized target so it has to be missiles.

Putting this further away, say typical MEO GPS-level orbit, solve so many problem, much cleaner orbit, much more illumination, much harder to reach and target, you just have to accept 100-200 ms latency.

 

[Dilandu]

On the other hand, the destruction can be done by any motivated and skilled decently sized team no matter the country.

This required such team to either have access to space-launch capabilities, or at least ASAT-capable missiles and equipment. While it's not impossible, it's sufficiently unlikely. And any non-government group THAT large and rich would clearly have better things to do than shooting down orbital data centers.

Basically any sounding rocket does the job of significantly damaging the array and particularly polluting its orbit.

It would require extremely precise calculations and decent tracking capability, to intercept orbiting object with sounding rocket. The size isn't as important as velocity; the slightest mistake of target movement calculation or the slighters deviation of the rocket would cause clear miss.

 

[Dilandu]

So this needs active space based defense, laser won't cut it at diverting the mass away from the kilometer-sized target so it has to be missiles.

Well, more like mines. A kinetic interceptor with very limited supply of delta-v, basically designed only to stand in the way of incoming missile.

 

[Michel Van]

more i read the comments here,
more becomes this project very doubtful...

 

[southwestforests]

Hey, let's see what else might be out there about orbital data centers,

https://www.cnbc.com/2024/06/27/europe-wants-to-deploy-data-centers-into-space-study-says.html

Within that,

The facilities that the study explored launching into space would orbit at an altitude of around 1,400 kilometers (869.9 miles) — about three times the altitude of the International Space Station. Dumestier explained that ASCEND would aim to deploy 13 space data center building blocks with a total capacity of 10 megawatts in 2036, in order to achieve the starting point for cloud service commercialization.

Each building block — with a surface area of 6,300 square meters — includes capacity for its own data center service and is launched within one space vehicle, he said.

In order to have a significant impact on the digital sector's energy consumption, the objective is to deploy 1,300 building blocks by 2050 to achieve 1 gigawatt, according to Dumestier.

Beyond the cloud: The concept of orbital data centres

Traditional data centres consume vast amounts of electricity. To address this issue, could the concept of orbital data centres be the answer?

datacentremagazine.com

The Space Review: Prospects for orbital data centers

That one has an interesting quote given some comments here,
(and why does the old Battletech miniatures game suddenly come to mind when talking thermal management?)

Terrestrial high-performance compute is heavy, primarily because of liquid thermal control systems. These need to be optimized. A continued reduction in space launch costs, coupled with advances in lightweight power and thermal management solutions, will be the key factors to watch in making orbital data centers practical.

Are Space-Based Data Centers the Future or Science Fiction?

In this blog, we'll explore the potential challenges and opportunities associated with space-based data centers

www.datacenters.com

 

[TomS]

i'm afraid not, there Data Center concept art has no radiators...

Having gone to actually read their prospectus, they do talk about cooling, and argue that they need about 1/3 the area for cooling as they use for solar power collection. I'm not sold on their math, but at least they are aware of the issue.

 

[publiusr]

Seeing how some aerospace collections that we thought were safe in libraries--I say charge the big ISPs, put only aerospace data in them--and make it free for all.

 

[Michel Van]

Scott Manley take on this

View: https://www.youtube.com/watch?v=d-YcVLq98Ew

 

[iverson]

In my opinion this thing does not seem very safe to preserve or exchange data, regardless of temperature changes it would need a very heavy radiation shielding... What is the advantage?

Exactly. Down here on earth, under the heavy shielding provided by the atmosphere, cosmic radiation already causes soft errors in data-storage hardware. We don't notice it all that often, because chips have built-in error-correction code (ECC). In essence, designers dedicate part of each byte to error checking. Typically, one of the eight bits in each byte is used as a parity bit. One bit in eight is usually enough to detect a single-bit error. This reduces the data rate a little but adds the redundancy needed for error detection (you really, really do not want undetected, "silent" data corruption).

In space, without the atmospheric shielding, you are probably going to see many more single-bit errors and, worse, more multi-bit errors that cannot be detected using a single parity bit. So I suspect that you'd need to dedicate much more of your available data storage and communications bandwidth to correcting errors, thus lowering performance. This is probably very acceptable in a massively over-resourced, experimental space program. But it does not seem like a competitive commercial proposition, especially now that most file storage and software applications are commodity priced and implemented using cloud architecture.

Cosmic radiation is a bigger deal than one might expect. Engineering once designed and released a critical memory chip without ECC in hopes of a performance gain. Within days, brand-new customer systems were panicking several times per day all over the world. I was at home on a Sunday. A very perturbed member of executive management actually called me to find out why customer CTOs were calling him.

 

[publiusr]

Perhaps this will help

Space travel: Protection from cosmic radiation with boron nitride nanotube fibers

With the success of the Nuri launch last year and the recent launch of the newly established Korea Aerospace Administration, interest in space has increased, and both the public and private sectors are actively investing in space-related industries such as space travel. However, exposure to...

phys.org

National lab takes its radiation expertise to space

An experiment designed to answer questions about the radiation environment for manned space missions was launched from Kennedy Space Center today.

phys.org

Starship could ease mass restraints by having some type of optical hardcopy (crystal memory) with radiation not able to destroy data and read/write completely.

No worry about a mob torching that library.

 

[iverson]

Perhaps this will help

Space travel: Protection from cosmic radiation with boron nitride nanotube fibers

With the success of the Nuri launch last year and the recent launch of the newly established Korea Aerospace Administration, interest in space has increased, and both the public and private sectors are actively investing in space-related industries such as space travel. However, exposure to...

phys.org

National lab takes its radiation expertise to space

An experiment designed to answer questions about the radiation environment for manned space missions was launched from Kennedy Space Center today.

phys.org

Starship could ease mass restraints by having some type of optical hardcopy (crystal memory) with radiation not able to destroy data and read/write completely.<snip>

I did not say that one could not provide shielding, though I doubt that one could match the protection provided by Earth's atmosphere and magnetic fields. The problem is economic rather than technical. Can one provide enough shielding and error-protection while hitting the price point for a terrestrial data center?

I spent half my working life, more less, in data centers. Data centers are not created for their own sake or even for the data's sake. They are cost centers in businesses that are always looking to reduce cost. This is why most on-premises data centers have now been outsourced to virtual server farms on huge AWS and Netflix clouds. The economies of scale are enough (or are perceived to be enough) to trump concerns about the security and availability of data in cloud implementations. The proposed orbital implementation adds enormous overhead that has nothing to do with data processing (earth-to-orbit transport, radiation shielding, much more severe environmental control, etc.). I doubt that the alleged energy savings can offset them.

The optical memory with which I am familiar is also not immune to radiation. For one thing, if you can write with light, light can change what you wrote. Stray light (including IR and RF) is everywhere. Plus, any physical material is subject to damage from high-energy particles like cosmic rays. So you still have to have enough redundancy (in the form of ECC or shielding) to cope with data loss when using optical media. Redundancy equals higher cost.

 

[iverson]

<snip>

National lab takes its radiation expertise to space

An experiment designed to answer questions about the radiation environment for manned space missions was launched from Kennedy Space Center today.

phys.org

<snip>

The radiation threat to astronauts is an example of my previous point. High-energy particles (cosmic rays, gamma rays, etc.) knock holes in our DNA, the biological equivalent of bits and bytes. The resulting errors in the genetic code cause mutations that are sometimes beneficial but are more often harmful or lethal to us (cancer) or our descendants.

Mutations are an issue even on Earth, with all that atmospheric and magnetic shielding above us. So evolution seems to have produced:

• error-correcting chemistry that can find and fix many of these errors
• a lot of redundant code that makes critical particle strikes less likely.

 

[Atomic Coyote]

Aside from the other issues with this, wouldn't drag and radiation pressure be a severe issue for an object with such massive solar panels in LEO?

What kind of littering fine are small Australian towns going to give when the solar maximum makes the data center crash into the outback 4 years early?