Search in 2D Matrix in DSA Java - Time & Space Complexity

Time Complexity: Search in 2D Matrix

O(rows * cols)

Understanding Time Complexity

We want to understand how the time to find a number in a 2D matrix grows as the matrix gets bigger.

Specifically, how many steps does it take to search through the matrix?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

public boolean searchMatrix(int[][] matrix, int target) {
    int rows = matrix.length;
    int cols = matrix[0].length;
    for (int i = 0; i < rows; i++) {
        for (int j = 0; j < cols; j++) {
            if (matrix[i][j] == target) {
                return true;
            }
        }
    }
    return false;
}
    

This code checks every element in the matrix one by one until it finds the target or finishes checking all elements.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Checking each element in the matrix.
• How many times: The outer loop runs once for each row, and the inner loop runs once for each column in that row.

How Execution Grows With Input

As the matrix size grows, the number of elements to check grows with the total number of rows times columns.

Input Size (rows x cols)	Approx. Operations
10 x 10	100 checks
100 x 100	10,000 checks
1000 x 1000	1,000,000 checks

Pattern observation: The number of checks grows quickly as the matrix gets bigger, multiplying rows and columns.

Final Time Complexity

Time Complexity: O(rows * cols)

This means the time to search grows directly with the total number of elements in the matrix.

Common Mistake

[X] Wrong: "Since the matrix is sorted, searching is always fast like O(log n)."

[OK] Correct: This code checks every element without using the sorted order to skip parts, so it still takes time proportional to all elements.

Interview Connect

Understanding how nested loops affect time helps you explain and improve search methods in real problems.

Self-Check

"What if we use binary search on each row instead of checking every element? How would the time complexity change?"

Practice

1. What is the main idea behind searching a target value in a 2D matrix where each row is sorted and the first integer of each row is greater than the last integer of the previous row?

easy

A. Treat the 2D matrix as a sorted 1D array and apply binary search

B. Search each row linearly until the target is found

C. Use a hash map to store all elements for quick lookup

D. Start from the bottom-left corner and move diagonally

Solution

1. Step 1: Understand matrix sorting property

   Each row is sorted and the first element of a row is greater than the last element of the previous row, so the entire matrix can be seen as a sorted list.

2. Step 2: Apply binary search on the matrix

   By treating the matrix as a 1D sorted array, binary search can be used efficiently to find the target.

3. Final Answer:

   Treat the 2D matrix as a sorted 1D array and apply binary search -> Option A

4. Quick Check:

   Matrix sorted row-wise = binary search [OK]

Hint: Think of the matrix as one sorted list for fast search [OK]

Common Mistakes:

• Searching each row linearly wastes time
• Using hash map wastes extra space
• Starting from bottom-left only works for different matrix types

2. Which of the following Java code snippets correctly calculates the middle index for binary search on a 2D matrix treated as a 1D array with total elements rows * cols?

easy

A. int mid = (low * high) / 2;

B. int mid = (low + high) / 2;

C. int mid = (low - high) / 2;

D. int mid = low + (high - low) / 2;

Solution

1. Step 1: Understand safe mid calculation

   Using low + (high - low) / 2 prevents integer overflow compared to (low + high) / 2.

2. Step 2: Check other options

   int mid = (low - high) / 2; and D do not correctly calculate mid index for binary search.

3. Final Answer:

   int mid = low + (high - low) / 2; -> Option D

4. Quick Check:

   Safe mid calculation = low + (high - low) / 2 [OK]

Hint: Use low + (high - low) / 2 to avoid overflow [OK]

Common Mistakes:

• Using (low + high) / 2 can cause overflow
• Using subtraction incorrectly for mid
• Multiplying low and high is wrong

3. Given the following Java code to search for target 5 in the matrix int[][] matrix = {{1,3,5,7},{10,11,16,20},{23,30,34,60}};, what will be the output?

public static boolean searchMatrix(int[][] matrix, int target) {
    int rows = matrix.length;
    int cols = matrix[0].length;
    int low = 0, high = rows * cols - 1;
    while (low <= high) {
        int mid = low + (high - low) / 2;
        int midValue = matrix[mid / cols][mid % cols];
        if (midValue == target) return true;
        else if (midValue < target) low = mid + 1;
        else high = mid - 1;
    }
    return false;
}

System.out.println(searchMatrix(matrix, 5));

medium

A. true

B. false

C. Compilation error

D. Runtime exception

Solution

1. Step 1: Understand matrix and target

   The matrix contains 5 at position (0,2). The binary search checks mid values correctly by converting 1D mid to 2D indices.

2. Step 2: Trace binary search steps

   Binary search finds 5 and returns true.

3. Final Answer:

   true -> Option A

4. Quick Check:

   5 found in matrix = true [OK]

Hint: Convert mid to row and col using mid/cols and mid%cols [OK]

Common Mistakes:

• Using wrong index conversion for mid
• Confusing row and column indices
• Returning false before full search

4. Identify the bug in the following Java method that searches a target in a 2D matrix where each row is sorted and the first integer of each row is greater than the last integer of the previous row:

public static boolean searchMatrix(int[][] matrix, int target) {
    int rows = matrix.length;
    int cols = matrix[0].length;
    int low = 0, high = rows * cols;
    while (low <= high) {
        int mid = low + (high - low) / 2;
        int midValue = matrix[mid / cols][mid % cols];
        if (midValue == target) return true;
        else if (midValue < target) low = mid + 1;
        else high = mid - 1;
    }
    return false;
}

medium

A. The mid calculation causes overflow

B. The high index should be rows * cols - 1, not rows * cols

C. The matrix indexing is incorrect

D. The loop condition should be low < high

Solution

1. Step 1: Check high initialization

   High should be the last valid index, which is rows * cols - 1, not rows * cols (which is out of bounds).

2. Step 2: Verify other parts

   Mid calculation and indexing are correct; loop condition low <= high is standard for binary search.

3. Final Answer:

   The high index should be rows * cols - 1, not rows * cols -> Option B

4. Quick Check:

   High index off by one causes error [OK]

Hint: High index = total elements - 1 for zero-based arrays [OK]

Common Mistakes:

• Setting high to rows * cols causes out-of-bounds
• Changing loop condition incorrectly
• Miscomputing mid index

5. You have a 2D matrix where each row is sorted in ascending order, but the first integer of each row is NOT necessarily greater than the last integer of the previous row. Which approach is best to search for a target value efficiently?

hard

A. Use binary search treating matrix as 1D sorted array

B. Perform binary search on each row separately

C. Start from top-right corner and move left/down based on comparison

D. Use linear search over all elements

Solution

1. Step 1: Understand matrix property

   Rows are sorted but no guarantee that next row's first element is greater than previous row's last element, so treating as 1D sorted array is invalid.

2. Step 2: Choose search method

   Starting from top-right corner, move left if current > target, move down if current < target, efficiently narrows search.

3. Final Answer:

   Start from top-right corner and move left/down based on comparison -> Option C

4. Quick Check:

   Top-right corner search = efficient for partially sorted matrix [OK]

Hint: Top-right corner search works when rows sorted but no global order [OK]

Common Mistakes:

• Applying 1D binary search wrongly
• Searching each row fully wastes time
• Using linear search is slow