Andy Adinets

Andy Adinets has been involved with GPU programming since 2005 and has been working at Nvidia as an AI Developer Technology Engineer for nearly 4 years, since the end of July 2017. He is currently working on a number of projects, including Forest Inference Library (FIL) and GPU sorting. He also likes working on optimization of various GPU algorithms when the opportunity arises.

Posts by Andy Adinets

Technical Walkthrough 0

Sparse Forests with FIL

The RAPIDS Forest Inference Library, affectionately known as FIL, dramatically accelerates inference (prediction) for tree-based models… 6 MIN READ
GPU Pro Tip
Technical Walkthrough 0

CUDA Pro Tip: Optimized Filtering with Warp-Aggregated Atomics

This post introduces warp-aggregated atomics, a useful technique to improve performance when many CUDA threads atomically update a single counter. 14 MIN READ
Technical Walkthrough 0

A CUDA Dynamic Parallelism Case Study: PANDA

Learn how Dynamic Parallelism of NVIDIA GPUs is being used to accelerate the discoveries of particle physics running on the PANDA experiment part of the Facility for Antiproton and Ion Research in Europe (FAIR). 11 MIN READ
Technical Walkthrough 0

CUDA Dynamic Parallelism API and Principles

This post is the second in a series on CUDA Dynamic Parallelism. In my first post, I introduced Dynamic Parallelism by using it to compute images of the… 13 MIN READ
Technical Walkthrough 0

Adaptive Parallel Computation with CUDA Dynamic Parallelism

Early CUDA programs had to conform to a flat, bulk parallel programming model. Programs had to perform a sequence of kernel launches, and for best performance… 13 MIN READ