The 2030 Computing Barrier: Solving Energy Overload

By 2030, the computing community will face an unprecedented crisis: the data we generate will outpace our ability to store and process it efficiently, while the energy demands of traditional computing will become unsustainable. The Landauer Limit—a fundamental barrier to energy efficiency—forces us to rethink how we compute. In this video, we dive into the upcoming computing bottleneck and explore groundbreaking solutions like reversable computing, protein-based machines, adiabatic computing, and DNA data storage. Discover how biotechnology and molecular systems could redefine energy-efficient computing and help avert the looming 2030 barrier.

00:00 - Intro
00:14 - Energy Problem
02:19 - Landauer Principle (or Limit)
04:31 - Reach Landaurer Limit in 2030
05:02 - adiabatic
06:48 - DNA Information Stores and Protein Computers
08:23 - Neuromorphic Computing
09:25 - NVIDIA/AMD
10:18 - Pre-Built PC Limits
11:34 - DLSS 4
13:07 - Wrapup

Follow me:
Twitter @djware55

[nozyy_]

interesting topic, thank you for the insight

[dezmondwhitney]

Nice video DJ. Not being well informed on this topic, it was very interesting to hear of the different ways in which improvements are being researched.
As a bit of a Luddite, I find certain aspects of A.I. and what may come along with it, concerning to me.

[IceBlade]

my bigest problem with DLSS is the amount of time it takes to generate those frames witch increas unput lag and those frames dose not inclued curent input of mouse/keyboard untill next real frame happens and the inputs now count again my second problem i would say is the lack of object permanance that AI dont understand so moving around causes alot of guesing of what is/was behind walls or static tings. and of course the extra artifacts you get.

[ttttol]

Given coal equivalent for data centers is short by 3 zeros.

[jholloway]

I'm just going to skip a few steps with organic computing and force the people in my basement to memorize 1's and 0's

[bertnijhof]

I''m fine, my Ryzen 5 5600GT has a 65W rating. It runs 90% of the time with a CPU load of ~10%. The CPU runs in power saver mode, because the 5600GT is overkill. I bought it in December and I never reached a 100% CPU load, also not during Windows 10/11 VM updates. My previous 2200G would easily reach 100% CPU load during Windows updates. The CPU temp = 45°C and normally I only use a 3rd gen nvme, a CPU and Case fan both running at ~1500 rpm. Typical usage would be <28W according to Google. The HDD is suspended after 5 minutes, the Monitor after 15 minutes and the PC after 30 minutes.

I run Ubuntu 24.04 LTS with OpenZFS and Virtualbox with 1 up to 4 VMs running, but with only 1 pair of hands :).
Most disk IO is done from the 8GB memory cache :) A typical cache hit rate is 98% or 99%.

Nice video, it made me check the power settings again and I corrected the default settings of the 2TB HDD, that I installed during the summer :)
So the Energy Overload is surely not my fault!

[Winnetou]

I don't understand that part at the start with the cost of delete vs write. Both in CPU registers, RAM and disk (be it solid state or HDD), a delete IS a write. Going over the special case of SSDs which have multiple bits per cell and writing would mean writing all 4 bits, the delete vs write doesn't make sense to me. At least, not in the current compute landscape.
And the "store the input so later you can simply switch, instead of delete" sounds like basic caching to me.

For consumer hardware, we are both getting more effiecient and not. If you look at smartphones and laptops, it is inarguable that we're getting much more efficient. And in general staying in the same power envelope, though the highend desktop replacement laptops are indeed more power hungry that what was 10 years ago.

On the desktop side... if we ignore a couple of generations from Intel (13th and 14th) then the CPUs I'd say are getting more efficient and also staying at a reasonable power draw, so same power envelope. Same for RAM and disks. It's the GPUs that are also more efficient, but have expanded, by quite a lot, the power envelope at mid and high end levels. But I would say that the raw hardware power is impressive.

On the datacenter side, 30, 000 tons of coal seems quite little. I expected something like 1 billion tons of coal. Funnily enough, a lot of electrity nowadays is consumed in AI. Feels like creating the problem in order to create the solution to me. Waaay too much desperate-ness in getting the AI upper hand is quite a clown show to me. I am expecting more and more regulations on AI as the data used is still highway robbery in most cases, and the energy used is just ludicrous, at least for the current and short-term future results. In the context of having to use less energy, so we can stop putting carbon into air.

Lastly on the prebuilt power limits or something similar. I don't know of having such a law, neither in EU nor in Romania where I live. However I do know that there is one for TVs (and other household electronic appliances, if I'm not mistaken) which actually limits the highend TVs quite a lot. Which, frankly, is quite stypid to me. If I get an 85" TV, you expect it to consume the same as a 40" inch one ? Not to mention that maybe I'm fully powered by my own solar panels. Who are you to decide that I can't use 200 more Watts for my TV ? On this theoretical setup, it would generate literally 0 extra carbon. And what's worse, because of this st00pid law, now people are incentivised to buy from abroad, which is worse for energy used (using energy to ship from the other side of the world instead of local) and worse for the economy (EU manufacturers cannot compete as well as those in other countries). Anyway, rant off.

[weinerdog]

LENR is real, mainstream cancelled it, over and over...with the support of TPTB....you gotta ask yourself "why?". If you answer it like i have, then you will see how difficult this is. As interesting as these details are, they are way down in the weeds. If thats what interests you, god bless and god speed.

[anasouardini]

Nuclear fusion FTW 😆
If only big coal and oil companies stop bullying the existing fission progress.

[stevensawolkin]

Hi, D.J. I quite enjoy all your videos. In the vast wasteland of linux lunkheads, ham 'n' eggers, and jingoists, you are an oasis of erudition, and twinkle-eyed wit. Perhaps you can answer a questio for me. I am keenly interested in (well, in everything, but...) linguistics and epistemology. Why, if you know, is "watt" pluralized, as in "550 watts", but "gig" is not, as in "400 gig"? Now, I am aware "gig" is a foreshortening of "gigabyte", or is it "gigabit"? But this does not, or ough tnot, impinge on the standard pluralization. Please remember that Sturgeon's Law suggests that the reply "It's just how they do it" is insufficient; as Ted would have me ask the NEXT question: Why is that just how they do it?" Thank you for your consideration of this "urgent" matter.

[jamerfunk]

The question is a false premise. consequently the Landauer 'limit, ' based upon yet another 'law' my personal hated, thermodynamics.
Recreating what was deleted is lunacy beyond words. It sounds straight outa paranoid city, how much energy is going to wasted doing that?
2/3's of the planet is covered in water, & the military is teleporting a Boeing 777 in midair on film.