Both are linear ADTs with O(1) ends, but the order of removal is opposite: a stack is LIFO (most recent first), a queue is FIFO (oldest first). Match the structure to the order your problem requires.
Stack (LIFO): push 1,2,3 -> pop order 3,2,1
Queue (FIFO): enq 1,2,3 -> deq order 1,2,3
| Use a STACK when... | Use a QUEUE when... |
|---|
| Undo/redo (latest action first) | Job scheduling (arrival order) |
| Browser back button | Print spooler / task queue |
| Function call/recursion | BFS / shortest path on a graph |
| Matching brackets, parsing | Message buffering between services |
| DFS (explore deepest first) | Rate limiting / request handling |
# DFS uses a stack: explore newest branch first
stack = [start]
while stack:
node = stack.pop() # LIFO -> goes deep
for nb in graph[node]: stack.append(nb)
# BFS uses a queue: explore nearest first
from collections import deque
q = deque([start])
while q:
node = q.popleft() # FIFO -> goes level by level
for nb in graph[node]: q.append(nb)
The LIFO-vs-FIFO choice directly changes algorithm behavior: the same graph traversal becomes DFS with a stack or BFS with a queue.
Recognizing whether your problem needs "most recent" or "oldest first" instantly points you to the right structure.