---
title: "CUB"
description: "1 posts about CUB, summarised, each linking to the original."
---

# CUB
> 1 posts about CUB, summarised, each linking to the original.

## Articles

### [The Modern CUDA Toolbox in Practice: A Step-by-Step Optimization Walkthrough](https://yomu.fyi/post/the-modern-cuda-toolbox-in-practice-a-step-by-step-optimization-walkth.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Elizabeth Goodman
- Published: Sep 2, 2026
- Type: Tutorial

An image-processing pipeline serves as the test case for a modern CUDA toolbox, starting with RGB-to-grayscale conversion, 32-by-32 tile median computation, and host/device transfers. It advances through six incremental changes: adopting CCCL and Compute Sanitizer for indexing bugs, adding NVTX for Nsight Systems benchmarks, replacing hand-rolled work with CUB algorithms, using pooled device containers and pinned host containers, and assigning each image its own stream with asynchronous copies. The implementation uses cuda::stream, cuda::copy\_bytes, cuda::device\_buffer, and CUB calls on per-thread streams, allowing kernels and transfers to overlap instead of executing sequentially on the default stream. The reported total duration falls from 6.8 seconds to about 23 milliseconds for three images, a roughly 300x improvement, while the post characterizes the code as safer and easier to maintain without low-level optimizations.