Skip to contentRU

Talk type: Talk

Производительность

Date: 23.05 / Start: 00:00 – Finish: 00:00

Speeding Up Synthesis: From TensorRT to CUDA C++

Talk in Russian

Presentation pdf

A talk about the CUDA C++ tool, which allows you to get the most out of the hardware you use.

#gpu
#tenzorrt
#algorithms

Speakers

Artyom Khoroshev
Company: SberDevices

Invited experts

Dmitry Izvolov

Other talks on «Производительность»
- Watch recording
  Talk type: Talk
  Room 2
  An Idea on Improving Multithreaded Code
  Alexander Kornilov
  Company: Kaspersky Lab
  Talk in Russian
- Watch recording
  Talk type: Talk
  Room 3
  The Practice of Tensor Kernel Programming
  Mikhail Lukin
  Company: Sudo
  Talk in Russian
- Watch recording
  Talk type: Talk
  Room 2
  GPU usage best practice from gamedev
  Rostislav Mikheev
  Company: Elverils
  Talk in Russian
- Watch recording
  Talk type: Talk
  Room 1
  GPU Driven Rendering Pipeline, or How Graphics are Written in Modern Video Games
  Eugene Grigorchuk
  Company: Elverils
  Talk in Russian
- Watch recording
  Talk type: Talk
  Teaching Cuckoo to Fly, or Concurrent Hash Map with Seqlocks
  Anton Potapov
  Company: Huawei
  Anton Malakhov
  Company: Huawei
  Talk in Russian
- Watch recording
  Talk type: Talk
  Advanced Linux Kernel Bypass Techniques in 2023
  Ashot Vardanian
  Company: Unum
  Talk in English
- Watch recording
  Talk type: Talk
  Room 3
  Performance Optimization Techniques
  Maksim Kita
  Company: Yandex
  Talk in Russian
- Watch recording
  Talk type: Talk
  Room 3
  Fastest Possible and the Most Flexible С++ Msgpack Codec
  Aleksander Lyapunov
  Company: VK
  Talk in Russian