Cache invalidation strategies in HLD - Architecture Diagram

System Overview - Cache invalidation strategies

This system explains how cache invalidation strategies work to keep cached data fresh and consistent with the main database. It focuses on common methods to update or remove stale data from the cache to ensure users get accurate information quickly.

Architecture Diagram

User
  |
  v
Load Balancer
  |
  v
API Gateway
  |
  v
Service Layer
  |
  +-------------------+
  |                   |
  v                   v
Cache <------------ Database
  |                   |
  +----> Message Queue <
       (for async updates)

Components

User

client

Initiates requests to the system

Load Balancer

load_balancer

Distributes incoming requests evenly to API Gateway instances

API Gateway

api_gateway

Handles client requests, routes them to appropriate services

Service Layer

service

Processes business logic, manages cache and database interactions

Cache

cache

Stores frequently accessed data for fast retrieval

Database

database

Stores the source of truth data

Message Queue

queue

Handles asynchronous cache invalidation or update messages

Request Flow - 11 Hops

User → Load Balancer

Load Balancer → API Gateway

API Gateway → Service Layer

Service Layer → Cache

Cache → Service Layer

Service Layer → Database

Database → Service Layer

Service Layer → Cache

Service Layer → User

Service Layer → Message Queue

Message Queue → Cache

Failure Scenario

Component Fails:Cache

Impact:Cache misses increase, causing more direct database queries and higher latency

Mitigation:System continues to serve data from database; cache rebuilds on demand; consider cache replication or fallback strategies

Architecture Quiz - 3 Questions

Which component is responsible for distributing incoming user requests evenly?

A Load Balancer

B API Gateway

C Cache

D Message Queue

Design Principle

This architecture demonstrates the importance of asynchronous cache invalidation to keep cached data fresh without blocking user requests, improving system responsiveness and scalability.

Practice

1. Which cache invalidation strategy automatically removes cached data after a set time period?

easy

A. Time-to-Live (TTL)

B. Write-through caching

C. Explicit invalidation

D. Cache-aside

Solution

1. Step 1: Understand TTL concept

   TTL means cached data expires automatically after a fixed time, ensuring freshness.

2. Step 2: Compare with other strategies

   Write-through updates cache on writes, explicit invalidation requires manual removal, cache-aside loads on demand.

3. Final Answer:

   Time-to-Live (TTL) -> Option A

4. Quick Check:

   Automatic expiry = TTL [OK]

Hint: TTL means cache expires after fixed time automatically [OK]

Common Mistakes:

• Confusing write-through with automatic expiry
• Thinking explicit invalidation is automatic
• Mixing cache-aside with TTL

2. Which of the following is the correct description of the write-through cache invalidation strategy?

easy

A. Cache is cleared manually by the application

B. Cache is updated only when data is read

C. Cache is updated immediately when data is written

D. Cache data expires after a fixed time

Solution

1. Step 1: Define write-through caching

   Write-through means every write to the database also updates the cache immediately.

2. Step 2: Eliminate other options

   Cache is updated only when data is read describes cache-aside, C is explicit invalidation, D is TTL.

3. Final Answer:

   Cache is updated immediately when data is written -> Option C

4. Quick Check:

   Write-through = update cache on write [OK]

Hint: Write-through updates cache on every write instantly [OK]

Common Mistakes:

• Confusing read update with write update
• Mixing explicit invalidation with write-through
• Assuming TTL applies to write-through

3. Consider a cache-aside strategy where the cache is empty initially. If a client requests data key K and the cache misses, what happens next?

medium

A. The system returns an error immediately

B. The system fetches data from the database and updates the cache

C. The system waits for explicit invalidation before fetching

D. The system uses stale data from cache

Solution

1. Step 1: Understand cache-aside behavior on miss

   Cache-aside loads data from the database on a cache miss and updates the cache.

2. Step 2: Check other options

   The system returns an error immediately is wrong because system does not error on miss, C is incorrect as explicit invalidation is unrelated, D is wrong because cache is empty.

3. Final Answer:

   The system fetches data from the database and updates the cache -> Option B

4. Quick Check:

   Cache miss triggers DB fetch and cache update [OK]

Hint: Cache-aside loads from DB on miss, then caches data [OK]

Common Mistakes:

• Assuming cache returns error on miss
• Confusing explicit invalidation with cache miss handling
• Thinking stale data is returned on miss

4. A system uses explicit invalidation but sometimes serves stale data. What is the most likely cause?

medium

A. Cache entries expire too quickly

B. Cache updates on every write immediately

C. Cache-aside strategy is used incorrectly

D. Invalidation messages are delayed or lost

Solution

1. Step 1: Analyze explicit invalidation issues

   Explicit invalidation depends on timely invalidation messages to remove stale cache entries.

2. Step 2: Identify cause of stale data

   If invalidation messages are delayed or lost, stale data remains in cache and is served.

3. Final Answer:

   Invalidation messages are delayed or lost -> Option D

4. Quick Check:

   Delayed invalidation causes stale cache [OK]

Hint: Delayed invalidation messages cause stale cache data [OK]

Common Mistakes:

• Thinking cache expiry too fast causes stale data
• Confusing write-through with explicit invalidation
• Assuming cache-aside errors cause stale data here

5. You design a high-traffic e-commerce system where product prices update rarely but must be accurate. Which cache invalidation strategy best balances freshness and performance?

hard

A. Use explicit invalidation triggered by price update events

B. Use write-through caching to update cache on every price change

C. Use cache-aside with manual cache clearing

D. Use a short TTL to expire cache frequently

Solution

1. Step 1: Analyze system requirements

   Prices update rarely but must be accurate, so cache should update only on changes, not frequently expire.

2. Step 2: Evaluate strategies

   Short TTL causes unnecessary reloads, write-through may be costly if writes are rare, cache-aside with manual clearing is error-prone.

3. Step 3: Choose explicit invalidation

   Explicit invalidation triggered by price update events ensures cache is fresh only when needed, balancing performance and accuracy.

4. Final Answer:

   Use explicit invalidation triggered by price update events -> Option A

5. Quick Check:

   Rare updates + accuracy = explicit invalidation [OK]

Hint: Rare updates + accuracy = explicit invalidation best [OK]

Common Mistakes:

• Choosing short TTL causing frequent reloads
• Assuming write-through is best for rare updates
• Ignoring event-driven invalidation benefits