AI Architecture & Systems
1
Microarchitecture
2
Kernel
3
Compiler
4
Architecture
5
System
6
Algorithms
Kernel
/
§2.7
Elementwise, Reduction, and Scan
2.7.1
Vector Add and Elementwise Fusion
#
2.7.2
Softmax and Online Softmax
#
2.7.3
LayerNorm, RMSNorm, and Reduction Layout
#
2.7.4
Prefix Sum, Scan, and Histogram
#
2.7.5
Top-k, Argmax, Sort, and Sampling
#
2.7.6
Welford's Algorithm and Stable Reduction
#
← Previous
2.6 Matrix Computation Across Shapes and Scenarios
→ Next
2.8 Fusion and Memory-Traffic Elimination