Tiling
التبليط
تقنية تقسّم المصفوفات الكبيرة إلى كتل صغيرة تتسع في الذاكرة السريعة (SRAM)، فتُعالَج كتلةً كتلة بدل تحميل المصفوفة بأكملها.
A technique that divides large matrices into small blocks that fit in fast memory (SRAM), processing them block by block instead of loading the entire matrix.
Also translated asالتقسيم إلى بلاطات، التجزئة الكُتَلية
First appears in this corpus in: FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness (2022)
Appears in these papers
- FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning2023in the sky ✦
- FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning2023in the sky ✦
- FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness2022in the sky ✦
- FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness2022in the sky ✦