Rik Kisnah - Blog

#GPU

Design an Inference API

Nov 11, 2025 Design an Inference API

OCI Powers America's AI Future at NVIDIA GTC 2025: Supercomputers, AI Factories, and Strategic Leadership

Oct 29, 2025 OCI Powers America's AI Future at NVIDIA GTC 2025: Supercomputers, AI Factories, and Strategic Leadership

From First Principles to Zettascale: How OCI's GPU/RDMA Architecture Redefines AI Infrastructure

Oct 26, 2025 From First Principles to Zettascale: How OCI's GPU/RDMA Architecture Redefines AI Infrastructure

Mixture of Experts and All-to-All Traffic

Sep 16, 2025 Mixture of Experts and All-to-All Traffic

Continuous Batching for Inference

Apr 15, 2025 Continuous Batching for Inference

Size the KV Cache

Feb 18, 2025 Size the KV Cache

Power and Cooling for GPU Racks

Aug 13, 2024 Power and Cooling for GPU Racks

Design a Burn-In Pipeline

Apr 9, 2024 Design a Burn-In Pipeline

Straggler Detection in Training Jobs

Jan 16, 2024 Straggler Detection in Training Jobs

Measure GPU Utilisation and MFU

Nov 14, 2023 Measure GPU Utilisation and MFU