Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.math > #646407

FYI: Unified Memory Architecture (UMA) (Re: Implementing Gas for a Compute Shader [Avoid TDR])

Cross-posted to 2 groups.

Show all headers | View raw

Hi,

These novel GPUs , that are part of AI Laptops,
feature Unified Memory Architecture (UMA).
In the case of my Ryzen the main memory is 32 GB,

and the GPU can access 16 GB. It is a design where
CPUs and GPUs share a single coherent memory space,
eliminating the need for separate host and

accelerator memory. Nevertheless the WebGPU API,
works with some copy and synchronization semantics
and buffer abstractions, which is not a big loss.

But all buffers reside in the same AI Lapop
RAM. And the MIMD architecture allows instructions
where seperate threads access the same

memory location, either for read or for write.
There is also a data type atomic(T), which
features operation such as AtomicAdd() etc.. etc..

BTW, I made a GitHub project of my exploration:

11.4 Giga Lips with a Budget Laptop
https://github.com/Jean-Luc-Picard-2021/gigabudget

BTW, pi-WAM is nevertheless optimized to have
no memory contention. On the other hand pi-WAM
is happy to access large memory areas.

Bye

BTW: The grandmother of these novel GPUs
is NVIDIAs Volta which already appeared in
2017, meanwhile we have 2026.

See also:

Starting with the NVIDIA Volta architecture,
Independent Thread Scheduling allows full
concurrency between threads, regardless of warp.
https://forums.developer.nvidia.com/t/back-to-simd/311983

The AMD Radeon 860M is an integrated graphics
processor that does not have its own dedicated
VRAM. Instead, it dynamically shares up to 50%
of your total system RAM with the CPU in a
standard Windows configuration.
https://www.amd.com/en/blogs/2025/faqs-amd-variable-graphics-memory-vram-ai-model-sizes-quantization-mcp-more.html

Ryann Likunov schrieb:
> Mild Shock wrote:
> 
>> Hi,
>>
>> I guess, you can read off how to do it here,
>> i.e. avoid TDR (DXGI_ERROR_DEVICE_HUNG 0x887A0006). The below compute
>> toys example has a similar
>>
>> approach, just like my π-WAM and Hack GPU backend, that is based on a
>> Instruction Set Architecture (ISA):
> 
> how come they are not able to map the local RAM as gpu arrays for using
> them as local AI resource
> 

Back to sci.math | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Re: Implementing Gas for a Compute Shader [Avoid TDR] (Re: AI dooms day escape: Güttinger Wald) Ryann Likunov <ka@ion.ru> - 2026-07-09 14:03 +0000
  FYI: Unified Memory Architecture (UMA) (Re: Implementing Gas for a Compute Shader [Avoid TDR]) Mild Shock <janburse@fastmail.fm> - 2026-07-09 19:41 +0200
    Re: FYI: Unified Memory Architecture (UMA) (Re: Implementing Gas for a Compute Shader [Avoid TDR]) Tanner Babadzhan <daz@atnt.ru> - 2026-07-09 20:15 +0000
      The Wider von Neumann Neck (Was: FYI: Unified Memory Architecture (UMA)) Mild Shock <janburse@fastmail.fm> - 2026-07-10 07:48 +0200
        The new MIMD Warp is the cherry on top (Re: The Wider von Neumann Neck) Mild Shock <janburse@fastmail.fm> - 2026-07-10 08:09 +0200
          Mac Neo: The dwarf with giant GPU muscles [macOS IOGPUFamily] (Re: The new MIMD Warp is the cherry on top) Mild Shock <janburse@fastmail.fm> - 2026-07-10 08:30 +0200
        Re: The Wider von Neumann Neck (Was: FYI: Unified Memory Architecture (UMA)) Landon Bahmutov <alaavh@ohbotlo.ru> - 2026-07-10 12:53 +0000
        Re: The Wider von Neumann Neck (Was: FYI: Unified Memory Architecture (UMA)) Olin Bagramov <laorb@ama.ru> - 2026-07-10 12:59 +0000
          Dumbwit, just run it on your RTX 5070 trash (Was: The Wider von Neumann Neck) Mild Shock <janburse@fastmail.fm> - 2026-07-10 15:44 +0200
            We are waiting Dumbwit: 1 Month, 3 Months, .. [node.js dawn] (Was: Dumbwit, just run it on your RTX 5070 trash) Mild Shock <janburse@fastmail.fm> - 2026-07-10 16:27 +0200
              Re: We are waiting Dumbwit: 1 Month, 3 Months, .. [node.js dawn] (Was: Dumbwit, just run it on your RTX 5070 trash) Russell Tuinov <ovul@ss.ru> - 2026-07-10 17:40 +0000
            Re: Dumbwit, just run it on your RTX 5070 trash (Was: The Wider von Neumann Neck) Lecel Bazulev <eccuz@ezelll.ru> - 2026-07-10 17:31 +0000
            We are still waiting for results! [Confused Dumbwit] (Was: Dumbwit, just run it on your RTX 5070 trash) Mild Shock <janburse@fastmail.fm> - 2026-07-10 22:44 +0200
              Re: We are still waiting for results! [Confused Dumbwit] (Was: Dumbwit, just run it on your RTX 5070 trash) Rian Babuh <baaaau@bbuaauba.ru> - 2026-07-10 21:26 +0000
                404 Brain not Found [GPU saturation] (Was: We are still waiting for results! [Confused Dumbwit]) Mild Shock <janburse@fastmail.fm> - 2026-07-10 23:30 +0200
                I nowhere talked about 10GB/s (Was: 404 Brain not Found [GPU saturation]) Mild Shock <janburse@fastmail.fm> - 2026-07-10 23:37 +0200
                Re: 404 Brain not Found [GPU saturation] (Was: We are still waiting for results! [Confused Dumbwit]) Wilfred Durmanov <roauw@odudv.ru> - 2026-07-10 21:44 +0000
                I nowhere said something about AI (Was: 404 Brain not Found [GPU saturation] (Was: We are still waiting for results! [Confused Dumbwit]) Mild Shock <janburse@fastmail.fm> - 2026-07-10 23:48 +0200
                Re: I nowhere said something about AI (Was: 404 Brain not Found [GPU saturation] (Was: We are still waiting for results! [Confused Dumbwit]) Junior Romagna <rr@rum.it> - 2026-07-10 21:59 +0000
                What do you not understand in "budget"? (Was: I nowhere said something about AI) Mild Shock <janburse@fastmail.fm> - 2026-07-11 00:02 +0200
                SWI makes only MLips not GLips (Was: What do you not understand in "budget"?) Mild Shock <janburse@fastmail.fm> - 2026-07-11 00:04 +0200
                Re: What do you not understand in "budget"? (Was: I nowhere said something about AI) Loman Barabashev <mahasm@aell.ru> - 2026-07-11 12:25 +0000
                micro penis got hurt by "budget" (Was: What do you not understand in "budget"?) Mild Shock <janburse@fastmail.fm> - 2026-07-12 19:17 +0200
                Re: What do you not understand in "budget"? (Was: I nowhere said something about AI) Ross Finlayson <ross.a.finlayson@gmail.com> - 2026-07-12 17:26 -0700
                herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Mild Shock <janburse@fastmail.fm> - 2026-07-13 06:51 +0200
                Re: herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Patryck Morozov <zzcm@mro.ru> - 2026-07-13 10:43 +0000
                nothing gets blocked except your brain [LPDDR5X-RAM] (Was: herpes blister rossy boy is confused) Mild Shock <janburse@fastmail.fm> - 2026-07-13 13:20 +0200
                flogging a dead horse PCI VRAM graphic cards (Was: nothing gets blocked except your brain [LPDDR5X-RAM]) Mild Shock <janburse@fastmail.fm> - 2026-07-13 13:51 +0200
                Re: nothing gets blocked except your brain [LPDDR5X-RAM] (Was: herpes blister rossy boy is confused) Roberto Bahmetov <tr@mrtt.ru> - 2026-07-13 14:20 +0000
                Re: herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Ross Finlayson <ross.a.finlayson@gmail.com> - 2026-07-13 07:48 -0700
                Re: herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Nel Baturov <orlu@bbeabuerln.ru> - 2026-07-13 15:16 +0000
                Re: herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Ross Finlayson <ross.a.finlayson@gmail.com> - 2026-07-13 10:56 -0700
                Re: herpes blister rossy boy is confused (Was: What do you not understand in "budget"?) Obelin Baisaroff <bbs@eieai.ru> - 2026-07-13 18:39 +0000
                GPUs evolved from SIMD to MIMD [NVIDIA Volta 2017] (Was: herpes blister rossy boy is confused) Mild Shock <janburse@fastmail.fm> - 2026-07-20 17:10 +0200
                Re: GPUs evolved from SIMD to MIMD [NVIDIA Volta 2017] (Was: herpes blister rossy boy is confused) Seldon Teterev <edne@vtt.ru> - 2026-07-20 16:14 +0000
                A.D.I.D.A.S. Gopnik lost his screws (Was: GPUs evolved from SIMD to MIMD [NVIDIA Volta 2017]) Mild Shock <janburse@fastmail.fm> - 2026-07-20 19:57 +0200

csiph-web