The RoarVM [1] is a research project that showed how to run Squeak Smalltalk on thousands of cores (at one point it ran on 10,000 cores).
I'm re-implementing it as a metacircular adaptive compiler and VM for a production operating system. We rewrite the STEPS research software and the Frank code [2] on a million core environment [3]. On the M4 processor we try to use all types of cores, CPU, GPU, neural engine, video hardware, etc.
$6500 depending on VAT. But 10-12 times M4 Mac mini's with 100 Gbps networking gives you triple the cores and 160 GB with 2.5 times the memeory bandwith if the sharding of the NN layers is done right.
$6500!! You may as well buy 5x3090s for $1000 each for 120GB ram, spend the extra $1500 on the sundries.
Like, I'm sure Nvidia is aware of Apple's "unified memory" as an alternative to their cards and yet...they aren't offering >24GB consumer cards yet, so clearly they don't feel threatened.
Don't get me wrong, I've always disliked Apple as a company, but the M series chips are brilliant, I'm writing this on one right now. But people seem to think that Apple will be able to get the same perf increases yoy when they're really stretching process limits by dumping everything onto the same die like that - where do they go from here?
That said Nvidia is using HBM so it does make me wonder why they aren't also doing memory on package with HBM, I think SK Hynix et al were looking at making this possible.
I'm glad we're headed in the direction of 3d silicon though, always seemed like we may as well scale in z, I imagine they can stack silicon/cooling/silicon/cooling etc. I'm sure they can use lithography to create cooling dies to sandwich between everything else. Then just pass connections/coolant through those.
As an Apple developer for 42 years I tried all permutations I could think of and found there is no way around the walled garden:
Apple Invites requires a mandatory iCloud+ account (minimal 0.99 euro/dollar per month) and a non-anonymous Apple ID requirement with credit card or bank account. It probably has a perpetual lock-in of your invite groups and tracking of all participants as well but I couldn't test this properly without paying 12 Euros.
In these comments there is a distinction between between activating the private personal Starlink accounts of user morphle and activating the Starlink accounts of other people by the same person on behalf of a rural internet coöperative in other countries.
We also helped rural Spain, rural Ukraine and on other continents and islands with setting up optical mesh and satellite internet, most of which had Starlink links as backup.
Our main project is an asynchronous wafer scale integration, a 1000 core asynchronous processors with 300 Megabyte of asynchronous SRAM (300 Kilobyte per core) and a huge Morphle Logic, our ‘asynchronous’ gate array.
A $600 supercomputer running Squeak Smalltalk and a GUI.
The second is a $5 10 core processor with 6 Megabyte memory for IoT, gigabit ethernet and Morphle Logic to compress video streams Squeak Smalltalk over ethernet. Much faster than a Raspberry PI 4 with a 4K camera, but not $100 but just $5 total cost.
> though more as a spectator than a participant in recent years,
That is a pity, I was hoping you would be a grad student or similar that could do a Masters or PhD thesis writing our software for the asynchronous manycore processor and the ‘asynchronous’ gate array.
Maybe you can help me locate another student who would do his masters with us?
Any volunteer would be helpful, it's just that we don't pay any salaries. But we do hand out samples......
I had a lecture on a larger version of this, a million core wafer scale integration running Smalltalk: https://vimeo.com/731037615
The bias in the "software engineering" roles you can search for is so strong that this search will work like a low pass filter, only bad programmers, ignorant coders and pop culture level will pass this search bias. 10X programmers, computer scientist and hackers will not find any job with this filter.
Maybe the makers of this search need to study what true programmers are like?
A few links to get started [1][2][3][4].
If you had restricted the search to Cobol, Microsoft or PHP your would have created the same type of bias against skilled programmers. https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
[1] There is still a war going on. Edsger Dijkstra.
what if you had access to just the data -- like a spreadsheet? and you could build your own search? also kind of don't like the search and want to rip it up
That is a perfect idea! Most skillfull hackers would indeed like to just search the raw data for a good job instead of relying on your precanned search.
I would happily pay for the raw data!
I would not pay for this search page.
I'm re-implementing it as a metacircular adaptive compiler and VM for a production operating system. We rewrite the STEPS research software and the Frank code [2] on a million core environment [3]. On the M4 processor we try to use all types of cores, CPU, GPU, neural engine, video hardware, etc.
We just applied for YC funding.
[1] https://github.com/smarr/RoarVM
[2] https://www.youtube.com/watch?v=f1605Zmwek8
[3] https://www.youtube.com/watch?v=wDhnjEQyuDk