← All parts of this equation

Equation 1 · Part 4 · From Origins to Frontier: A History of AI Memory Systems and the Bandwidth Wall

multiplication

compute(t)bandwidth(t)  =  compute(0)bandwidth(0)⋅(3.01.6)t/2,\frac{\text{compute}(t)}{\text{bandwidth}(t)} \;=\; \frac{\text{compute}(0)}{\text{bandwidth}(0)} \cdot \left(\frac{3.0}{1.6}\right)^{t/2},
multiplication

What this part means

Multiply the quantities on either side.

Its job in the formula

Multiply the quantities on either side.

The passage around this formula

The scale of that abstract argument, made concrete decades later, is stark. Gholami and colleagues, surveying twenty years of server hardware, report that peak hardware FLOPS scaled at roughly 3.0 times every two years while DRAM bandwidth scaled at only 1.6 times and interconnect bandwidth at 1.4 times over the same interval [ 13 ] . Those are compounding rates, and compounding rates diverge violently over long periods. If the two multipliers applied uniformly across a full twenty years — an extrapolation the source itself does not perform, offered here only to size the shape of the problem, not as a number the cited paper states — the ratio between available arithmetic and available…

Read this part in the article →

Learn the underlying idea

Multiplication scales one quantity by another. A dot, a cross, or adjacent symbols can indicate a product.

Open the illustrated multiplication: combining factors guide →

Sources cited in the surrounding passage

These citations provide research context; check each source for the exact claim it supports.