Modular: AI’s compute fragmentation: what matrix multiplication teaches us
We are building a next-generation AI developer platform for the world. Read our latest post on how AI’s compute fragmentation: what matrix multiplication teaches us
Modular: AI’s compute fragmentation: what matrix multiplication teaches us Qualcomm to Acquire Modular. Read More → March 23, 2023 AI’s compute fragmentation: what matrix multiplication teaches us Eric Johnson Abdul Dakkak Chad Jarvis Engineering Compute fragmentation is holding AI back AI is powered by a virtuous circle of data, algorithms (“models”), and compute. Growth in one pushes needs in the others and can grossly affect the developer experience on aspects like usability and performance. Today, we have more data and more AI model research than ever before, but compute isn’t scaling at t
related reading
- Modular: The world's fastest unified matrix multiplicationmodular.com
- AI Chip Architecturesjacobpeake.com
- How to Optimize a CUDA Matmul Kernel for cuBLAS-like Performance: a Worklogsiboehm.com
- Inside NVIDIA GPUs: Anatomy of high performance matmul kernels - Aleksa Gordićaleksagordic.com
- AI Chip Architecturesjepeake.com
- Navigating the High Cost of AI Compute | Andreessen Horowitza16z.com
- Strangely, Matrix Multiplications on GPUs Run Faster When Given "Predictable" Data! [short]thonking.ai
- Computational Power and AI - AI Now Instituteainowinstitute.org
- There Are No New Ideas in AI… Only New Datasetsblog.jxmo.io
- Transformers Inference Optimization Toolset | AstraBlogastralord.github.io
- NVIDIA Tensor Core Evolution: From Volta To Blackwellnewsletter.semianalysis.com
- AI progress is about to speed up | Epoch AIepoch.ai