Study plan — ذاكر إيه وبعدين حل إيه
| # | Slides to study | Topics | Then solve |
|---|---|---|---|
| 1 | L1: 1–36 L2: 5–20 | Why multicore · global sum · task vs data · classifications | Flashcards (L1, L2) |
| 2 | L2: 21–40 | Flynn · shared vs distributed · UMA / NUMA / COMA | Tutorial 1 Q1–Q5, Q7 |
| 3 | L3: 7–15 | Speedup · efficiency · Amdahl | Tutorial 1 Q6 + Lab 1 |
| 4 | L3: 16–47 | Hierarchy · cache basics · direct · associative | Tutorial 2 Q1 |
| 5 | L3: 48–69 | Set-associative · replacement · read/write policies | Tutorial 2 Q2–Q4 + cache simulator |
| 6 | L4: 3–18 | Multiprocessor variations · coherence problem · write protocols | Tutorial 2 Q5–Q7 |
| 7 | L4: 19–25 | Directory · snooping · MESI | Tutorial 2 Q8–Q10 + MESI tracer |
| 8 | L5: 1–40 L6: 3–18 | CPU vs GPU · CUDA model · host API · kernel launch | Lab 2 + Drills D7–D10 |
| 9 | L6: 19–40 | 2D grids · picture kernel · matmul · scalability | Drills D11–D14 |
| 10 | Lab 1 code | pthreads modifications | Drills D1–D6 + all flashcards |