The Serving Stack
Start hereCollectionsNewsletter
Sign in

cuda

2 posts

GPU FOR INFERENCE · 0148 SMs · 24 MiB L2 · 273 GB/s DRAM
gpu-for-inferencegpucuda

A GPU in One Picture: SMs, Warps, and Memory

The parts of a GPU that decide inference speed, from streaming multiprocessors to DRAM, with one real GPU's numbers and three measurements.

by The Serving Stack
CUDA FOR INFERENCE · 01xyy[i] = a·x[i] + y[i], ONE THREAD EACH235 GB/s, 86% OF THE MEMORY
cuda-for-inferencecudagpu

Your First CUDA Kernel, Compiled and Profiled

Write SAXPY in CUDA, compile it for a Blackwell GPU, check it, time it, and read what Nsight Systems and Nsight Compute say about it.

by The Serving Stack

Inference Notes

Once a week: what changed in serving AI models, and what it means if you build, run or pay for them, in about five minutes.

The Serving Stack

Inference Engineering, In Depth

Navigate

HomeStart hereCollectionsNewsletterAll articles

Collections

Inference EnginesHow an LLM Request Becomes TokensLLMs, ExplainedThe GPU for Inference EngineersQuantization in PracticeServing in Production

Links

RSS Feed

Legal

Privacy PolicyTerms of ServiceCookie Policy
© 2026 The Serving Stack. All rights reserved.