Why streams handle event logs in Redis - Performance Analysis

Time Complexity: Why streams handle event logs

O(k)

Understanding Time Complexity

When using Redis streams to handle event logs, it's important to understand how the time it takes to process events changes as the log grows.

We want to know how the cost of adding and reading events grows with the number of events stored.

Scenario Under Consideration

Analyze the time complexity of adding and reading events in a Redis stream.

XADD mystream * event "login" user "alice"
XADD mystream * event "logout" user "bob"
XRANGE mystream - + COUNT 10
XREAD COUNT 5 STREAMS mystream 0
    

This code adds events to a stream and reads a range or a count of events from it.

Identify Repeating Operations

Look for repeated actions that affect performance.

• Primary operation: Adding events with XADD and reading events with XRANGE or XREAD.
• How many times: Each event is added once; reading can request multiple events at once.

How Execution Grows With Input

Adding an event takes about the same time no matter how many events are in the stream.

Reading events takes time proportional to how many events you ask for, not the total stream size.

Input Size (n)	Approx. Operations
10 events	Adding: constant time; Reading 5 events: about 5 operations
100 events	Adding: constant time; Reading 10 events: about 10 operations
1000 events	Adding: constant time; Reading 100 events: about 100 operations

Pattern observation: Adding events stays fast regardless of stream size; reading scales with how many events you request.

Final Time Complexity

Time Complexity: O(k)

This means the time depends on the number of events you read or write at once, not the total number stored.

Common Mistake

[X] Wrong: "Adding events gets slower as the stream grows because it has to check all previous events."

[OK] Correct: Redis streams add events efficiently without scanning old events, so adding stays fast even with many events.

Interview Connect

Understanding how Redis streams handle event logs helps you explain efficient data handling in real systems, showing you know how to manage growing data smoothly.

Self-Check

"What if we read events without specifying a count limit? How would that affect the time complexity?"

Practice

1. Why are Redis streams well-suited for handling event logs?

easy

A. Because they only store one event at a time

B. Because they delete old events automatically

C. Because they store events in order with unique IDs

D. Because they require complex setup to use

Solution

1. Step 1: Understand Redis streams storage

   Redis streams store events sequentially with unique IDs, preserving order.

2. Step 2: Match this to event log needs

   Event logs need ordered, unique entries to track actions accurately.

3. Final Answer:

   Because they store events in order with unique IDs -> Option C

4. Quick Check:

   Streams = Ordered unique events [OK]

Hint: Streams keep events ordered with unique IDs [OK]

Common Mistakes:

• Thinking streams delete old events automatically
• Believing streams store only one event
• Assuming streams need complex setup

2. Which Redis command is used to add a new event to a stream?

easy

A. XREAD

B. XADD

C. XRANGE

D. XDEL

Solution

1. Step 1: Identify command to add events

   XADD is the Redis command designed to add new entries to a stream.

2. Step 2: Differentiate from other commands

   XREAD reads events, XRANGE queries a range, XDEL deletes entries.

3. Final Answer:

   XADD -> Option B

4. Quick Check:

   XADD = Add event [OK]

Hint: XADD adds events to streams [OK]

Common Mistakes:

• Confusing XADD with XREAD or XRANGE
• Thinking XDEL adds events
• Using XRANGE to add events

3. Given the Redis stream commands below, what will XRANGE mystream - + return?

XADD mystream * event "login" user "alice"
XADD mystream * event "logout" user "bob"

medium

A. All events in 'mystream' from the oldest to newest

B. Only the first event added to 'mystream'

C. An error because XRANGE needs a count parameter

D. No events because the stream is empty

Solution

1. Step 1: Understand XRANGE command

   XRANGE with '-' and '+' returns all entries from oldest to newest in the stream.

2. Step 2: Check stream content after XADD commands

   Two events were added: a login by alice and a logout by bob.

3. Final Answer:

   All events in 'mystream' from the oldest to newest -> Option A

4. Quick Check:

   XRANGE - + = all events [OK]

Hint: XRANGE - + returns all stream events [OK]

Common Mistakes:

• Thinking XRANGE returns only one event
• Believing XRANGE requires a count parameter
• Assuming stream is empty after XADD

4. You run the command XADD mystream * event but get an error. What is the likely cause?

medium

A. XRANGE command is required before XADD

B. Using * instead of a specific ID

C. Stream 'mystream' does not exist

D. Missing field-value pairs after the event name

Solution

1. Step 1: Check XADD syntax requirements

   XADD requires field-value pairs after the ID; 'event' lacks a value.

2. Step 2: Validate other options

   Using '*' is correct for auto ID, streams can be created automatically, and XRANGE is unrelated.

3. Final Answer:

   Missing field-value pairs after the event name -> Option D

4. Quick Check:

   XADD needs field-value pairs [OK]

Hint: XADD needs pairs: field and value [OK]

Common Mistakes:

• Thinking * is invalid for ID
• Assuming stream must exist before XADD
• Believing XRANGE must run first

5. You want to track user actions in a Redis stream but only keep the last 100 events to save memory. Which command option helps achieve this?

hard

A. XADD mystream MAXLEN ~ 100 * event "login" user "alice"

B. XADD mystream * event "login" user "alice" COUNT 100

C. XADD mystream * event "login" user "alice" MAXLEN 100

D. XADD mystream * event "login" user "alice" LIMIT 100

Solution

1. Step 1: Understand MAXLEN option in XADD

   MAXLEN limits stream length; '~' makes it approximate for better performance.

2. Step 2: Identify correct syntax for trimming

   XADD mystream MAXLEN ~ 100 * event "login" user "alice" uses 'MAXLEN ~ 100' which trims stream to about 100 entries efficiently.

3. Final Answer:

   XADD mystream MAXLEN ~ 100 * event "login" user "alice" -> Option A

4. Quick Check:

   MAXLEN ~ 100 trims stream length [OK]

Hint: Use MAXLEN ~ 100 to trim stream length efficiently [OK]

Common Mistakes:

• Using MAXLEN without ~ causes slower trimming
• Using COUNT or LIMIT which are invalid here
• Not trimming stream and causing memory issues