[HN Gopher] The von Neumann bottleneck is impeding AI computing?
       ___________________________________________________________________
        
       The von Neumann bottleneck is impeding AI computing?
        
       Author : Nezteb
       Score  : 25 points
       Date   : 2025-09-26 21:12 UTC (1 hours ago)
        
 (HTM) web link (research.ibm.com)
 (TXT) w3m dump (research.ibm.com)
        
       | mwkaufma wrote:
       | The old saw from corporations that want to sell you an locked-
       | down alternative to general-purpose computing -- now for "AI"
        
         | bahmboo wrote:
         | Huh, I did not get that from the article. The main takeaway for
         | me was doing ALU operations in memory resulting in massive
         | energy savings. There is still a von Neumann architecture
         | running the show.
        
       | nyrikki wrote:
       | Nit,
       | 
       | ARM processors primarily use a modified Harvard architecture,
       | including the raspberry pi pico.
        
         | NooneAtAll3 wrote:
         | this isn't about Harvard/VonNeuman split/no-split between
         | i-cache and d-cache
         | 
         | I think this post is more about... compute in memory? if I got
         | it right?
        
           | danudey wrote:
           | Sort of? It's about locality of data; this has often been a
           | bottleneck, which is why we have CPU caches to keep data
           | extremely close to the CPU cores with practically zero
           | latency and throughput limitations compared to fetching from
           | main memory. Unfortunately now we're shuffling terabytes of
           | data through our algorithms and the CPU spends a huge amount
           | of its time waiting for the next batch of data to come in
           | through the pipe.
           | 
           | This is, IIRC, part of why Apple's M-series chips are as
           | performant as they are: they not only have a unified memory
           | architecture which eliminates the need to copy data from CPU
           | main memory to GPU or NPU main memory to operate on it (and
           | then copy the result back) but the RAM being on the package
           | means that it's slightly "more local" and the memory channels
           | can be optimized for the system they're going to be connected
           | to.
        
         | bobmcnamara wrote:
         | Nit: RP2040 is a Von Neumann. There's only one AHB port on the
         | m0.
         | 
         | Edit: see also ARM7TDMI, Cortex-m0/0+/1, and probably a few
         | others. All the big stuff is modified Harvard or very rarely
         | pure Harvard.
        
       | Animats wrote:
       | Actual result: "This new process promises to increase the number
       | of optical fibers that can be connected at the edge of a chip, a
       | measure known as beachfront density, by six times."
       | 
       | Faster interconnects are always nice, but this is more like
       | routine improvement.
        
         | bahmboo wrote:
         | "In recent inference tests run on a 3-billion-parameter LLM
         | developed from IBM's Granite-8B-Code-Base model, NorthPole was
         | 47 times faster than the next most energy-efficient GPU and was
         | 73 times more energy efficient than the next lowest latency
         | GPU."
         | 
         | It's also fascinating that they are experimenting with analog
         | memory because it pairs so well with model weights
        
       | lomase wrote:
       | Why they don't use AI to create a new architecture?
        
         | observationist wrote:
         | https://github.com/GAIR-NLP/ASI-Arch
         | 
         | This is being done, with great results so far. As models get
         | better, architecture search and creation and refinment
         | improves, driving a reinforcement loop. At some point in the
         | near future the big labs will likely start seeing significant
         | returns from methods like this, translating into better and
         | faster AI for consumers.
        
           | lomase wrote:
           | I skimed the repo and only found sloop. Can you point out
           | where I can find those new architectures you talk about?
        
         | jedberg wrote:
         | Do you want SkyNet? That's how you get SkyNet.
        
           | amelius wrote:
           | No, that's how you get SlopNet.
        
       | owyn wrote:
       | If you follow the press release rabbit a few clicks, there's an
       | article in Science describing the architecture in more detail:
       | 
       | https://www.science.org/doi/full/10.1126/science.adh1174
       | 
       | Also they've been working on this for 10+ years so it's not
       | exactly new news.
        
       | stego-tech wrote:
       | IBM initially leads with the more salient point (current
       | architecture designs are hindering frontier computing concepts),
       | then just kinda...relents into iterative improvement.
       | 
       | Which is fine! I am all for iterative improvements, it's how we
       | got to where we are today. I just wish more folks would start
       | openly admitting that our current architecture designs are
       | broadly based off "low hanging fruit" of early electronics and
       | microprocessors, followed by a century of iterative improvements.
       | With the easy improvements already done and universally
       | integrated, we're stuck at a crossroads:
       | 
       | * Improve our existing technologies iteratively and hope we break
       | through some barrier to achieve rapid scaling again
       | 
       | OR
       | 
       | * Accept that we cannot achieve new civilizational uplifts with
       | existing technologies, and invest more capital into frontier R&D
       | (quantum processing, new compute substrates, etc)
       | 
       | I feel like our current addiction to the AI CAPEX bubble is a
       | desperate Hail Mary to validate our current tech as the only way
       | forward, when in fact we haven't really sufficiently explored
       | alternatives in the modern era. I could very well be wrong, but
       | that's the read I get from the hardware side of things and
       | watching us backslide into the 90s era of custom chips to achieve
       | basic efficiency gains again.
        
       ___________________________________________________________________
       (page generated 2025-09-26 23:00 UTC)