Tagged articles
3 articles
Page 1 of 1
Baidu Tech Salon
Baidu Tech Salon
Apr 2, 2025 · Artificial Intelligence

PaddlePaddle Framework 3.0 Released: Five Core Innovations for Large Models and Scientific Computing

PaddlePaddle 3.0, launched on April 1 2025, introduces five core innovations—including dynamic‑static unified automatic parallelism, a training‑inference integrated PIR, high‑order automatic differentiation for scientific computing, a one‑stage CINN compiler, and heterogeneous multi‑chip adaptation—that dramatically reduce distributed‑training code, boost performance up to four‑fold, and extend the framework to aerospace, automotive, meteorology and life‑science applications while remaining fully compatible with the 2.0 API.

Deep LearningPaddlePaddleautomatic parallelism
0 likes · 21 min read
PaddlePaddle Framework 3.0 Released: Five Core Innovations for Large Models and Scientific Computing
Architects' Tech Alliance
Architects' Tech Alliance
Sep 1, 2024 · Industry Insights

How Heterogeneous Hardware Is Shaping the Future of System Software

The article analyzes the rise of heterogeneous hardware as a new trend, explains the systemic challenges caused by the gap between single‑purpose hardware capabilities and diverse application needs, and outlines two research pathways—hardware aggregation and application‑driven feedback—illustrated with concrete examples such as GPUs, RDMA, NVM, HTM, and TEE/SGX.

application feedbackhardware aggregationheterogeneous hardware
0 likes · 6 min read
How Heterogeneous Hardware Is Shaping the Future of System Software
JD Cloud Developers
JD Cloud Developers
Aug 2, 2024 · Operations

How JD’s Advertising Platform Optimizes Load Balancing for Heterogeneous Clusters

Exploring the evolution of JD’s advertising online model system, this article examines the challenges of heterogeneous hardware load balancing, outlines static and dynamic strategies—including DNS, Nginx, LVS, Ribbon, and Dubbo—and presents a multi‑objective framework that improves service availability and resource utilization, achieving up to 20%+ efficiency gains.

Distributed Systemsheterogeneous hardwareload balancing
0 likes · 17 min read
How JD’s Advertising Platform Optimizes Load Balancing for Heterogeneous Clusters