corpus.blog
56,966 blogs · 4,417,066 posts Claim your blog
Record · corpus.blog/posts/01a08778-8d56-70a2-8927-97c72f59b7c1

Domain specific architectures for AI inference

fleetwood.dev · published 8 March 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at fleetwood.dev Wayback capture
2blogs citing
First cited
1 Dec 2025
Most recent
1 Dec 2025
Rank this month
not ranked
Links from this post
84
Words captured
6,653

Cited by

Blog
In the post
Date
considerthebulldog.com
Touching the Elephant - TPUs Read at the source ↗
1 Dec 2025
Suriya's site - Technical Writing & Life
to learn Read at the source ↗
—

Links from this post

Link
Host
https://en.wikipedia.org/wiki/Moore%27s_law
en.wikipedia.org
https://en.wikipedia.org/wiki/Dennard_scaling
en.wikipedia.org
https://en.wikipedia.org/wiki/Domain-specific_architecture
en.wikipedia.org
https://people.inf.ethz.ch/omutlu/pub/onur-DAC-LightningTalk-MemoryCentricComputing-13-July-2023.pdf
people.inf.ethz.ch
https://x.com/vrushankdes
x.com
https://horace.io/brrr_intro.html
horace.io
https://en.wikipedia.org/wiki/Adder_(electronics
en.wikipedia.org
https://en.wikipedia.org/wiki/Multiple_buffering
en.wikipedia.org
https://en.wikipedia.org/wiki/Direct_memory_access
en.wikipedia.org
https://pytorch.org/blog/hopper-tma-unit/
pytorch.org
https://en.wikipedia.org/wiki/Memory
en.wikipedia.org
https://en.wikipedia.org/wiki/Scratchpad_memory
en.wikipedia.org
https://en.wikipedia.org/wiki/Static_random-access_memory
en.wikipedia.org
https://cerebras.ai/
cerebras.ai
https://groq.com/
groq.com
https://tenstorrent.com/en
tenstorrent.com
https://en.wikipedia.org/wiki/Roofline_model
en.wikipedia.org
https://jax-ml.github.io/scaling-book/inference/
jax-ml.github.io
https://huggingface.co/docs/transformers/v4.49.0/en/perf_train_gpu_many
huggingface.co
https://en.wikipedia.org/wiki/Multilayer_perceptron
en.wikipedia.org
https://jax-ml.github.io/scaling-book/training/
jax-ml.github.io
https://github.com/deepseek-ai/DeepSeek-V3/blob/592fd5daf8177b205af11651bbb31a1834a8b0e0/inference/configs/config_671B.json
github.com
https://arxiv.org/pdf/2412.19437v1
arxiv.org
Learning to reason with LLMs Read at the source ↗
openai.com
https://en.wikipedia.org/wiki/Thinking,_Fast_and_Slow
en.wikipedia.org
https://en.wikipedia.org/wiki/Exploration%E2%80%93exploitation_dilemma
en.wikipedia.org
TPUv5e: The New Benchmark in Cost-Efficient Inference and Training for Read at the source ↗
semianalysis.com
https://en.wikipedia.org/wiki/Systolic_array
en.wikipedia.org
https://en.wikipedia.org/wiki/Colossus_computer
en.wikipedia.org
https://gwern.net/timing
gwern.net
https://arxiv.org/pdf/2304.01433
arxiv.org
https://www.reuters.com/technology/openai-set-finalize-first-custom-chip-design-this-year-2025-02-10/
reuters.com
https://tenstorrent.com/
tenstorrent.com
Introducing Triton: Open-source GPU programming for neural networks Read at the source ↗
openai.com
https://en.wikipedia.org/wiki/Jim_Keller_(engineer
en.wikipedia.org
https://www.youtube.com/watch?v=lPX1H3jW8ZQ
youtube.com
https://www.barrons.com/authors/tae-kim
barrons.com
https://x.com/firstadopter/status/1691877797487165443
x.com
AI Server Cost Analysis – Memory Is The Biggest Loser Read at the source ↗
semianalysis.com
https://www.dramexchange.com/
dramexchange.com
https://en.wikipedia.org/wiki/Flynn%27s_taxonomy
en.wikipedia.org
https://s201.q4cdn.com/141608511/files/doc_financials/2025/q3/ed2a395c-5e9b-4411-8b4a-a718d192155a.pdf
s201.q4cdn.com
https://blogs.microsoft.com/on-the-issues/2025/01/03/the-golden-opportunity-for-american-ai/
blogs.microsoft.com
https://youtu.be/Qgd3OK5DZWI?si=PRlwlb8AWfkQ0MvW&t=288
youtu.be
https://www.youtube.com/watch?v=Qgd3OK5DZWI
youtube.com
https://en.wikipedia.org/wiki/Systems_neuroscience
en.wikipedia.org
https://youtu.be/Qgd3OK5DZWI?si=PDlMbShiOG4hH1bX&t=1809
youtu.be
https://www.youtube.com/watch?v=MC223HlPdK0
youtube.com
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9373921
ieeexplore.ieee.org
https://arxiv.org/pdf/2109.14320
arxiv.org
https://arxiv.org/pdf/2007.00072
arxiv.org
https://www.youtube.com/watch?v=nHHsYp7ZkHQ
youtube.com
https://www.youtube.com/watch?v=XkgtANeDrm8
youtube.com
https://docs.nvidia.com/deeplearning/performance/dl-performance-gpu-background/index.html
docs.nvidia.com
https://www.telesens.co/2018/07/26/understanding-roofline-charts/
telesens.co
https://www.telesens.co/2018/07/30/systolic-architectures/
telesens.co
https://x.com/Tim_Dettmers/status/1856338240099221674
x.com
https://www.corsix.org/
corsix.org
https://developer.nvidia.com/blog/nvidia-hopper-architecture-in-depth/
developer.nvidia.com
https://docs.nvidia.com/deeplearning/performance/pdf/GPU-Performance-Background-User-Guide.pdf
docs.nvidia.com
https://jax.readthedocs.io/en/latest/pallas/tpu/pipelining.html
jax.readthedocs.io
https://iitd-plos.github.io/col729/lec/matrix_multiplication.html
iitd-plos.github.io
https://arxiv.org/pdf/2411.04330
arxiv.org
https://www.nextplatform.com/2024/02/27/he-who-can-pay-top-dollar-for-hbm-memory-controls-ai-training/
nextplatform.com
https://old.hotchips.org/hc31/HC31_T3_Cloud_TPU_Codesign.pdf
old.hotchips.org
https://epoch.ai/gradient-updates/how-has-deepseek-improved-the-transformer-architecture
epoch.ai
https://le.qun.ch/en/blog/2023/05/13/transformer-batching/
le.qun.ch
https://arxiv.org/pdf/2205.14135
arxiv.org
https://arxiv.org/ftp/arxiv/papers/2304/2304.01433.pdf
arxiv.org
https://dl.acm.org/doi/10.1145/1022594.1022596
dl.acm.org
https://arxiv.org/pdf/2208.11174
arxiv.org
https://gwern.net/doc/ai/scaling/hardware/2021-jouppi.pdf
gwern.net
https://arxiv.org/pdf/1911.05289
arxiv.org
https://arxiv.org/pdf/0712.3329
arxiv.org
https://jax-ml.github.io/scaling-book
jax-ml.github.io
https://kipp.ly/blog/transformer-inference-arithmetic/
kipp.ly
https://arxiv.org/abs/2412.19437
arxiv.org
https://arxiv.org/pdf/2405.04434
arxiv.org
https://arxiv.org/pdf/2309.06180
arxiv.org
https://pubmed.ncbi.nlm.nih.gov/23298528/
pubmed.ncbi.nlm.nih.gov
https://www.usenix.org/system/files/osdi22-yu.pdf
usenix.org
https://arxiv.org/abs/1704.04760
arxiv.org
https://www.eecs.harvard.edu/~htk/publication/1982-kung-why-systolic-architecture.pdf
eecs.harvard.edu
https://pubmed.ncbi.nlm.nih.gov/28728020/
pubmed.ncbi.nlm.nih.gov