Skip to main content

21. Number of Islands

mediumAsked at ByteDance

Count connected land regions in a grid — ByteDance uses it to test BFS/DFS scaffolding before scaling to user-cluster detection for recommendations.

By Alex Chen, Founder, InterviewChamp.AI · Last verified

Problem

Given an m x n grid of '1' (land) and '0' (water), return the number of islands. An island is formed by orthogonally connecting land cells; assume the grid is bordered by water.

Constraints

  • 1 <= m, n <= 300
  • grid[i][j] is '0' or '1'

Examples

Example 1

Input
grid = [["1","1","0"],["1","1","0"],["0","0","1"]]
Output
2

Example 2

Input
grid = [["1","0","1"],["0","0","0"],["1","0","1"]]
Output
4

Approaches

1. Union-find every land neighbor

Treat each land cell as a node, union with orthogonal land neighbors, count distinct roots.

Time
O(m * n * α)
Space
O(m * n)
// build DSU over m*n cells, union orthogonal land neighbors, count roots

Tradeoff:

2. DFS flood fill

Scan the grid. On each unvisited land cell, increment the count and flood-fill its connected region in place.

Time
O(m * n)
Space
O(m * n)
function numIslands(grid) {
  const m = grid.length, n = grid[0].length;
  let count = 0;
  const dfs = (r, c) => {
    if (r < 0 || c < 0 || r >= m || c >= n || grid[r][c] !== '1') return;
    grid[r][c] = '0';
    dfs(r + 1, c); dfs(r - 1, c); dfs(r, c + 1); dfs(r, c - 1);
  };
  for (let r = 0; r < m; r++) {
    for (let c = 0; c < n; c++) {
      if (grid[r][c] === '1') { count++; dfs(r, c); }
    }
  }
  return count;
}

Tradeoff:

ByteDance-specific tips

ByteDance interviewers want you to flag the recursion-depth risk for 300x300 grids and offer an iterative BFS variant — the same defensive move their backend team makes in production crawlers.

Solve it now

Free. No sign-up. Python and JavaScript run instantly in your browser.

Output

Press Run or Cmd+Enter to execute

Practice these live with InterviewChamp.AI

Drill Number of Islands and other ByteDance interview questions under real-loop conditions with instant feedback on your reasoning, complexity claims, and code.

Practice these live with InterviewChamp.AI →