Explore machine learning concepts related to parallelism. Clear explanations and practical insights.
Flynn's Classification explained — SISD, SIMD, MISD, MIMD with interactive architecture explorer, SIMD evolution from MMX to AMX, branch divergence visualization, and workload-architecture throughput comparison.
Complete guide to CUDA MPS — architecture, performance benchmarks vs time-slicing and MIG, thread percentage planning, production deployment with systemd and Kubernetes, profiling with nsys, and troubleshooting.
A CUDA context is a per-device container of GPU state; a CUDA stream is an in-order execution queue inside a context. They answer different questions and operate at different lifetimes — confusing them is one of the most common ways CUDA code ends up slow or unsafe.
A CUDA stream is a queue of GPU operations that execute in order. Understanding streams is the difference between a GPU at 30% utilization and one running flat out — they are how kernels and memory copies overlap on real hardware.
Explore the concept of CUDA contexts, their role in managing GPU resources, and how they enable parallel execution across multiple CPU threads.