Bitmap indexes in DBMS Theory - Time & Space Complexity

Time Complexity: Bitmap indexes

O(n)

Understanding Time Complexity

Bitmap indexes help speed up searching in databases by using bits to represent data presence.

We want to understand how the time to search grows as the data size increases.

Scenario Under Consideration

Analyze the time complexity of using a bitmap index to find rows matching a condition.

-- Assume a bitmap index on a column with distinct values
SELECT ROWID FROM table WHERE bitmap_index & condition_bitmap = condition_bitmap;
    

This code uses bitwise AND on bitmap indexes to quickly find matching rows.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Bitwise AND on bitmap segments representing rows.
• How many times: Once per bitmap segment, which covers many rows at once.

How Execution Grows With Input

As the number of rows grows, the bitmap size grows proportionally, so more bits must be processed.

Input Size (n rows)	Approx. Operations (bitwise checks)
10	10 bits processed
100	100 bits processed
1000	1000 bits processed

Pattern observation: The work grows linearly with the number of rows because each row corresponds to one bit.

Final Time Complexity

Time Complexity: O(n)

This means the time to search grows directly in proportion to the number of rows in the table.

Common Mistake

[X] Wrong: "Bitmap indexes always give constant time search regardless of data size."

[OK] Correct: Even though bit operations are fast, the bitmap size grows with data, so more bits must be checked as data grows.

Interview Connect

Understanding how bitmap indexes scale helps you explain database search efficiency clearly and confidently.

Self-Check

What if the bitmap index was compressed? How would that affect the time complexity?

Practice

1. What is the main advantage of using bitmap indexes in a database?

easy

A. They improve write performance for frequent updates

B. They reduce the size of the database tables

C. They automatically backup the database

D. They speed up queries on columns with few unique values

Solution

1. Step 1: Understand the purpose of bitmap indexes

   Bitmap indexes are designed to speed up searches on columns that have a small number of distinct values, like gender or status.

2. Step 2: Compare options with this purpose

   Only They speed up queries on columns with few unique values correctly states that bitmap indexes speed up queries on such columns. Other options describe unrelated features.

3. Final Answer:

   They speed up queries on columns with few unique values -> Option D

4. Quick Check:

   Bitmap indexes = speed up queries on low-cardinality columns [OK]

Hint: Bitmap indexes help when column values repeat often [OK]

Common Mistakes:

• Thinking bitmap indexes improve write speed
• Confusing bitmap indexes with data compression
• Assuming bitmap indexes backup data automatically

2. Which of the following is the correct way to describe a bitmap index?

easy

A. An index that stores row pointers in a tree structure

B. An index that stores full copies of data rows

C. An index that uses bits to represent the presence of values

D. An index that hashes values for quick lookup

Solution

1. Step 1: Recall bitmap index structure

   Bitmap indexes use bits (0 or 1) to indicate whether a row contains a specific value in a column.

2. Step 2: Match description to options

   An index that uses bits to represent the presence of values correctly describes this bit-based representation. Other options describe different index types.

3. Final Answer:

   An index that uses bits to represent the presence of values -> Option C

4. Quick Check:

   Bitmap index = bit representation of data presence [OK]

Hint: Bitmap means bits show if value exists in row [OK]

Common Mistakes:

• Confusing bitmap indexes with B-tree indexes
• Thinking bitmap indexes store full data rows
• Assuming bitmap indexes use hashing

3. Consider a column Gender with values 'M' or 'F' in a table of 1000 rows. How does a bitmap index represent this data?

medium

A. Two bitmaps, one for 'M' and one for 'F', each with 1000 bits

B. One bitmap with 2000 bits combining both values

C. A bitmap with 1000 bits storing the actual characters 'M' or 'F'

D. A bitmap that stores row numbers where values appear

Solution

1. Step 1: Understand bitmap index for low-cardinality columns

   For each distinct value, a bitmap index creates a bitmap with one bit per row indicating presence (1) or absence (0).

2. Step 2: Apply to 'Gender' column

   Since 'Gender' has two values ('M' and 'F'), there will be two bitmaps, each 1000 bits long, one for 'M' and one for 'F'.

3. Final Answer:

   Two bitmaps, one for 'M' and one for 'F', each with 1000 bits -> Option A

4. Quick Check:

   Bitmap per distinct value = two bitmaps of 1000 bits [OK]

Hint: One bitmap per distinct value, bits equal to rows [OK]

Common Mistakes:

• Thinking bitmap stores actual characters
• Assuming one combined bitmap for all values
• Confusing bitmap with row number storage

4. A bitmap index is used on a column that is frequently updated. What is the likely problem?

medium

A. Bitmap indexes improve update speed but slow down queries

B. Bitmap indexes slow down updates and cause locking issues

C. Bitmap indexes automatically adjust to frequent updates without issues

D. Bitmap indexes convert updates into batch inserts

Solution

1. Step 1: Recall bitmap index behavior on updates

   Bitmap indexes are not ideal for columns with frequent updates because changing bits can cause locking and slow performance.

2. Step 2: Evaluate options

   Bitmap indexes slow down updates and cause locking issues correctly states that bitmap indexes slow down updates and cause locking. Other options are incorrect or misleading.

3. Final Answer:

   Bitmap indexes slow down updates and cause locking issues -> Option B

4. Quick Check:

   Frequent updates + bitmap index = slow and locking [OK]

Hint: Bitmap indexes bad for frequent updates [OK]

Common Mistakes:

• Assuming bitmap indexes speed up updates
• Thinking bitmap indexes handle updates automatically
• Believing bitmap indexes convert updates to inserts

5. In a data warehouse, a bitmap index is created on a Region column with 5 distinct values. Which scenario best explains why bitmap indexes are preferred here?

hard

A. The column has few unique values and queries often filter by region

B. The column is updated every minute with new region codes

C. The column stores large text descriptions of regions

D. The column has millions of unique region codes

Solution

1. Step 1: Identify characteristics suitable for bitmap indexes

   Bitmap indexes work best on columns with few unique values and mostly read-only data, like in data warehouses.

2. Step 2: Analyze each option

   The column has few unique values and queries often filter by region fits perfectly: few unique values and frequent filtering. Options B, C, and D describe unsuitable scenarios.

3. Final Answer:

   The column has few unique values and queries often filter by region -> Option A

4. Quick Check:

   Few unique values + read-heavy queries = bitmap index ideal [OK]

Hint: Bitmap indexes suit low-cardinality, read-heavy columns [OK]

Common Mistakes:

• Using bitmap indexes on frequently updated columns
• Applying bitmap indexes to high-cardinality columns
• Confusing bitmap indexes with full-text indexes