← All parts of this equation

Equation 2 · Part 5 · Locality Is the Whole Game: The Memory Hierarchy and What a Kernel Does Not Read

≤

P  ≤  min⁡(Pmax⁡,  I⋅B),P \;\le\; \min\left(P_{\max},\; I \cdot B\right),
≤

What this part means

Less than or equal to.

Its job in the formula

Less than or equal to.

The passage around this formula

The quantity that decides which of these limits binds is arithmetic intensity : the number of arithmetic operations a kernel performs per byte it moves across a given level. Williams, Waterman and Patterson formalised the consequence as the roofline model, which relates attainable floating-point performance to operational intensity and memory bandwidth in a single visual bound [ 5 ] . Written as an inequality, attainable performance P satisfies P  ≤  min⁡(Pmax⁡,  I⋅B)P \;\le\; \min\left(P_{\max},\; I \cdot B\right). with Pmax⁡P_{\max} the peak arithmetic rate, I the arithmetic intensity in operations per byte, and B the achievable bandwidth at the level that dominates. The two terms cross at a ridge point I∗I^{*} = Pmax⁡P_{\max}/B . Below it a kernel is…

Read this part in the article →

Learn the underlying idea

An inequality compares values without claiming they are equal. It describes a range, threshold, or bound that a quantity may satisfy.

Open the illustrated inequalities: bounds and allowed ranges guide →

Sources cited in the surrounding passage

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