17 docs tagged with "patterns"

Backtracking searches a space of candidate solutions by building them one choice at a time, and

Modern CMake Philosophy

Divide and conquer breaks a problem into independent subproblems of the same kind, solves those

Dynamic programming applies when a problem has overlapping subproblems — the same sub-computation

A singly-linked list, or anything shaped like one — a permutation's i -> p[i] mapping, a

A greedy algorithm makes the choice that looks best right now and never reconsiders it. When that

A calendar full of meetings, a set of (start, end) ranges to merge, a question like "how many

"For each element, find the nearest element to the right that is bigger" looks like it needs a nested

Almost every GPU kernel, however specialized, is built from a small set of recurring data-access shapes. Recognizing which pattern a problem is — before writing any code — tells you how parallelizable it is, what its likely performance limiter will be, and often points directly at a library implementation that already exists and is already tuned. This page names those shapes once, and every later applied-kernel page in this knowledge base assumes you already know these names — "this is a reduction" or "this needs a scan" is meant to carry full meaning by the time you reach folder 13.

The pattern controls the line around your message — timestamp, level, logger name, thread id — and

Summing a range of an array costs time proportional to the range. Do it once and nobody notices; do it

The named algorithms in the earlier sections are instances of a smaller number of recurring

This page is a reference, not a tutorial — see Problem-Solving Patterns Overview for a

Dynamic Programming names the idea — overlapping subproblems plus

Test Doubles Taxonomy

"Find the k largest" looks like a sorting problem, and sorting solves it — but sorting also computes

Both patterns replace a nested loop with a single pass by maintaining two indices that only ever move