Arithmetic intensity = FLOPs / Bytes accessed
Image: Cancella, CC BY-SA 4.0, via Wikimedia Commons
Arithmetic intensity = FLOPs / Bytes accessed
instruction-level parallelism (ILP) achieves: multiple operations per clock cycle
ILP = average number of instructions per clock cycle
Delay-line memory
CPU speed grows faster than memory speed
Flashbulb memory
Flashbulb memories are vivid but not always accurate
Matrix multiplication algorithm
Ever wondered how computers speed up multiplying huge numbers?
Dynamic random-access memory
DRAM requires periodic refreshing to maintain data integrity
quantization to INT8 doubles throughput
Quantization to INT8 doubles throughput because tensor cores process INT8 2x faster
Swipe through 100 ML concepts daily
Open Pocket Polymath