Bigtable schema design in GCP - Step-by-Step Execution

Process Flow - Bigtable schema design

Identify access patterns

↓

Design row key for fast lookup

↓

Choose column families

↓

Decide on column qualifiers

↓

Plan for data versioning and TTL

↓

Implement schema in Bigtable

Start by understanding how data will be accessed, then design row keys and column families accordingly, finally implement the schema in Bigtable.

Execution Sample

GCP

Row key: userID#timestamp
Column families: info, activity
Columns: info:name, info:email, activity:page_views
TTL: 30 days

This schema stores user info and activity with a composite row key for efficient time-based queries.

Process Table

Step	Action	Details	Result
1	Identify access patterns	Need to query user activity by user and time	Access pattern clear
2	Design row key	Use userID#timestamp to allow range scans by user	Row key format set
3	Choose column families	Separate 'info' and 'activity' for logical grouping	Column families defined
4	Decide column qualifiers	info:name, info:email, activity:page_views	Columns chosen
5	Plan TTL	Set TTL to 30 days to expire old activity	TTL configured
6	Implement schema	Create table with above design in Bigtable	Schema ready
7	Exit	Schema design complete	Stop

💡 All design steps completed, schema ready for use

Status Tracker

Variable	Start	After Step 2	After Step 3	After Step 4	After Step 5	Final
row_key	undefined	userID#timestamp	userID#timestamp	userID#timestamp	userID#timestamp	userID#timestamp
column_families	none	none	info, activity	info, activity	info, activity	info, activity
columns	none	none	none	info:name, info:email, activity:page_views	info:name, info:email, activity:page_views	info:name, info:email, activity:page_views
TTL	none	none	none	none	30 days	30 days

Key Moments - 3 Insights

Why do we use a composite row key like userID#timestamp?

Using userID#timestamp allows efficient queries for all activities of a user in time order, as shown in execution_table step 2.

Why separate data into column families like 'info' and 'activity'?

Column families group related data and help with performance and access control, as explained in execution_table step 3.

What is the purpose of setting TTL in Bigtable schema?

TTL automatically deletes old data to save space and keep data fresh, as planned in execution_table step 5.

Visual Quiz - 3 Questions

Look at the execution_table, what is the row key format after step 2?

A timestamp#userID

B userID

C userID#timestamp

D timestamp

💡 Hint

Check the 'Details' column in execution_table row with Step 2

At which step are the column families defined?

A Step 2

B Step 3

C Step 1

D Step 4

💡 Hint

Look at the 'Action' column in execution_table for when column families are chosen

If TTL was not set, what would change in the variable_tracker?

A TTL would remain 'none' after all steps

B TTL would be set to 30 days anyway

C TTL would be set to 0 days

D TTL would be set to 60 days

💡 Hint

Check the TTL row in variable_tracker to see how TTL changes after step 5

Concept Snapshot

Bigtable schema design steps:
1. Identify how data is accessed
2. Design row keys for fast lookups
3. Group columns into families
4. Choose column qualifiers
5. Set TTL for data expiration
6. Implement schema in Bigtable

Full Transcript

Bigtable schema design starts by understanding how you will access your data. Then you create a row key that supports those access patterns, often combining identifiers like userID and timestamp. Next, you organize your data into column families to group related columns, which helps with performance and management. You decide on specific columns inside those families. You also plan for data lifecycle by setting TTL to automatically delete old data. Finally, you implement this design in Bigtable. This step-by-step approach ensures efficient and scalable data storage.