Contribute to zsc/ai_compiler_tutorial development by creating an account on GitHub.
本章深入探讨 NUMA(Non-Uniform Memory Access,非统一内存访问)架构下的 AI 编译器优化技术。随着 200T 参数级模型的出现,单一计算节点已无法满足计算和内存需求,NUMA 架构成为高性能计算的必然选择。本章将从 NUMA 基础概念出发,详细讨论亲和性设置、本地内存 ...
Abstract: Due to the increasing threats from possible large-scale quantum computers, post-quantum cryptography (PQC) has drawn significant attention from various communities recently. In particular, ...
China is exploring new approaches to reduce its reliance on Nvidia’s CUDA software, which plays a key role in the company’s ...
Toilet brushes, cutting boards and cleaning products don’t last forever. Here’s how to know when to toss these and other household items.
It’s available for free on both Android and iOS, so there’s really no barrier to giving it a try. If you’re curious about ...
Generic formats like JSON or XML are easier to version than forms. However, they were not originally intended to be human-readable but machine-readable. Since many applications require a ...