jackfiled
462fbb28ac
Some checks failed
Build blog docker image / Build-Blog-Image (push) Failing after 14s
Signed-off-by: jackfiled <xcrenchangjun@outlook.com> Reviewed-on: #21
81 lines
2.2 KiB
Markdown
81 lines
2.2 KiB
Markdown
---
|
|
title: High Performance Computing 25 SP Heterogeneous Computing
|
|
date: 2025-05-10T00:36:20.5391570+08:00
|
|
tags:
|
|
- 高性能计算
|
|
- 学习资料
|
|
---
|
|
|
|
|
|
Heterogeneous Computing is on the way!
|
|
|
|
<!--more-->
|
|
|
|
## GPU Computing Ecosystem
|
|
|
|
CUDA: NVIDIA's Architecture for GPU computing.
|
|
|
|
|
|
|
|
## Internal Buses
|
|
|
|
**HyperTransport**:
|
|
|
|
Primarily a low latency direct chip to chip interconnect, supports mapping to board to board interconnect such as PCIe.
|
|
|
|
**PCI Expression**
|
|
|
|
Switched and point-to-point connection.
|
|
|
|
**NVLink**
|
|
|
|
|
|
|
|
**OpenCAPI**
|
|
|
|
Heterogeneous computing was in the professional world mostly limited to HPC, in the consumer world is a "nice to have".
|
|
|
|
But OpenCAPI is absorbed by CXL.
|
|
|
|
## CPU-GPU Arrangement
|
|
|
|
|
|
|
|
#### First Stage: Intel Northbrige
|
|
|
|
|
|
|
|
### Second Stage: Symmetric Multiprocessors:
|
|
|
|
|
|
|
|
### Third Stage: Nonuniform Memory Access
|
|
|
|
And the memory controller is integrated directly in the CPU.
|
|
|
|
|
|
|
|
So in such context, the multiple CPUs is called NUMA:
|
|
|
|
|
|
|
|
And so there can be multi GPUs:
|
|
|
|
|
|
|
|
### Fourth Stage: Integrated PCIe in CPU
|
|
|
|
|
|
|
|
And there is such team *integrated CPU*, which integrated a GPU into the CPU chipset.
|
|
|
|
|
|
|
|
And the integrated GPU can work with discrete GPUs:
|
|
|
|
|
|
|
|
### Final Stage: Multi GPU Board
|
|
|
|
|