C.W.K.
Stream
Quiz · 5 questions

🔢 Matrices — Where Tiling Earns Its Keep

Coalesced access, reduction, naive vs tiled GEMM, the matrix_ops CLI

Level 0Beginner
0 XP0/38 lessons0/12 achievements
0/100 XP to next level100 XP to go0% complete

Quiz

01What is tiling in GPU computing?
Hint
It's about reuse, not just chopping work up.
02What is a reduction operation?
Hint
Think softmax / LayerNorm / loss.
03What are bank conflicts in shared memory?
Hint
Why does padding sometimes speed up shared-memory kernels?
04How do you convert (row, col) to a flat index in a row-major matrix?
Hint
ld is the stride between rows.
05How much faster is tiled GEMM vs naive GEMM in this track's RTX 4090 demo?
Hint
Look at the table in the lesson — naive vs tiled both in TFLOP/s.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.