Pub/sub limitations (no persistence) in Redis - Time & Space Complexity

Time Complexity: Pub/sub limitations (no persistence)

O(n)

Understanding Time Complexity

When using Redis Pub/Sub, messages are sent to subscribers instantly without saving them.

We want to understand how the time to deliver messages grows as more subscribers join.

Scenario Under Consideration

Analyze the time complexity of this Redis Pub/Sub publish operation.

# Subscribe clients to a channel
SUBSCRIBE channel1

# Publish a message to all subscribers
PUBLISH channel1 "Hello subscribers!"
    

This code sends a message to all clients subscribed to channel1 immediately without storing it.

Identify Repeating Operations

Look for repeated actions in the publish process.

• Primary operation: Sending the message to each subscriber.
• How many times: Once for each subscriber connected to the channel.

How Execution Grows With Input

As the number of subscribers grows, the work to send messages grows too.

Input Size (n)	Approx. Operations
10 subscribers	10 sends
100 subscribers	100 sends
1000 subscribers	1000 sends

Pattern observation: The number of sends grows directly with the number of subscribers.

Final Time Complexity

Time Complexity: O(n)

This means the time to deliver a message grows linearly with the number of subscribers.

Common Mistake

[X] Wrong: "Publishing a message takes the same time no matter how many subscribers there are."

[OK] Correct: Each subscriber must get the message separately, so more subscribers mean more work.

Interview Connect

Understanding how message delivery time grows helps you design systems that handle many users efficiently.

Self-Check

"What if Redis stored messages for offline subscribers? How would that change the time complexity of publishing?"

Practice

1. What happens to messages sent via Redis Pub/sub if no clients are subscribed at that moment?

easy

A. Messages are lost and not stored anywhere.

B. Messages are saved and delivered later when clients subscribe.

C. Messages are queued until a client connects.

D. Messages are stored in a Redis list automatically.

Solution

1. Step 1: Understand Redis Pub/sub behavior

   Redis Pub/sub delivers messages only to clients currently subscribed.

2. Step 2: Analyze message persistence

   Since Pub/sub does not save messages, if no client listens, messages are lost.

3. Final Answer:

   Messages are lost and not stored anywhere. -> Option A

4. Quick Check:

   Pub/sub = no persistence [OK]

Hint: Pub/sub delivers live only, no message storage [OK]

Common Mistakes:

• Assuming messages queue up for later delivery
• Thinking messages are saved automatically
• Confusing Pub/sub with Redis streams

2. Which of the following is the correct Redis command to subscribe to a channel named news?

easy

A. SUBSCRIBE TO news

B. PUBLISH news

C. SUBSCRIBE news

D. LISTEN news

Solution

1. Step 1: Recall Redis Pub/sub subscribe syntax

   The correct command to subscribe is SUBSCRIBE <channel>.

2. Step 2: Match command to channel name

   Using SUBSCRIBE news subscribes to the channel named 'news'.

3. Final Answer:

   SUBSCRIBE news -> Option C

4. Quick Check:

   Subscribe syntax = SUBSCRIBE channel [OK]

Hint: Subscribe uses 'SUBSCRIBE channel' command [OK]

Common Mistakes:

• Using PUBLISH instead of SUBSCRIBE
• Adding extra words like TO
• Using LISTEN which is not a Redis command

3. Consider this Redis Pub/sub scenario:

Client A subscribes to channel 'chat'.
Client B publishes message 'Hello' to 'chat'.
Client C subscribes to 'chat' after the message is published.

Which clients receive the message 'Hello'?

medium

A. Only Client A

B. Clients A and C

C. Only Client C

D. Clients A, B, and C

Solution

1. Step 1: Identify subscribers at publish time

   Only Client A was subscribed when Client B published the message.

2. Step 2: Understand Pub/sub delivery

   Messages are sent only to clients subscribed at publish time; Client C subscribed later and misses the message.

3. Final Answer:

   Only Client A -> Option A

4. Quick Check:

   Subscribers at publish = receivers [OK]

Hint: Only current subscribers get messages [OK]

Common Mistakes:

• Assuming late subscribers get past messages
• Including publisher as receiver
• Thinking all clients get all messages

4. You wrote this Redis Pub/sub code snippet:

SUBSCRIBE news
PUBLISH news "Update 1"

But your subscriber never receives the message. What is the likely problem?

medium

A. The PUBLISH command syntax is incorrect.

B. The subscriber must be subscribed before the message is published.

C. Redis does not support Pub/sub messaging.

D. The channel name 'news' is reserved and cannot be used.

Solution

1. Step 1: Check Pub/sub message delivery timing

   Subscribers must be listening before messages are published to receive them.

2. Step 2: Analyze the code order

   If PUBLISH runs before SUBSCRIBE completes, subscriber misses the message.

3. Final Answer:

   The subscriber must be subscribed before the message is published. -> Option B

4. Quick Check:

   Subscribe first, then publish [OK]

Hint: Subscribe before publish to receive messages [OK]

Common Mistakes:

• Thinking PUBLISH syntax is wrong
• Believing Redis lacks Pub/sub support
• Assuming channel names are restricted

5. You want to ensure messages are not lost if no clients are currently subscribed. Which Redis feature should you use instead of Pub/sub?

hard

A. Use Redis transactions to save messages.

B. Use Pub/sub with multiple channels.

C. Increase Redis memory to save messages.

D. Redis Streams to store and consume messages later.

Solution

1. Step 1: Understand Pub/sub limitation

   Pub/sub does not store messages; they are lost if no subscriber is present.

2. Step 2: Identify Redis feature for persistence

   Redis Streams allow storing messages so clients can read them later.

3. Step 3: Evaluate other options

   Using multiple channels or transactions does not provide message persistence; increasing memory does not save messages automatically.

4. Final Answer:

   Redis Streams to store and consume messages later. -> Option D

5. Quick Check:

   Use Streams for persistence, not Pub/sub [OK]

Hint: Use Redis Streams for message persistence [OK]

Common Mistakes:

• Trying to fix Pub/sub by adding channels
• Thinking transactions save Pub/sub messages
• Assuming more memory stores messages