Algocraft / System Design / Caching

⚡

Caching

Dramatically reduce latency by storing frequently accessed data in fast memory

Low Latency Read Scaling Redis / Memcached CDN

🎮 What you can do

Set cache capacity and eviction policy (LRU / LFU / FIFO)
Pick an access pattern (Random, Hot-spot, Sequential)
Adjust simulation speed

👁 What to watch

Cache grid highlights on hit (green) and eviction (red flash)
Hit-rate gauge climbs as the cache warms up
Average latency drops as hit rate improves

Cache Capacity

Eviction Policy

Access Pattern

Simulation Speed

Hit Rate

0%

0

Hits

0

Misses

Cache Tier Simulation ● Stopped

💻

Client

Last Key

—

⚡ Cache (RAM)

0/8

1ms access time

🗄️

Database

100ms access time

Queries

0

Idle

RECENT REQUESTS

0

Total Requests

0

Evictions

0ms

Avg Latency

0ms

Time Saved

Key Concepts

🎯 When to Use

Read-heavy workloads (>80% reads)
Expensive DB queries that repeat
Session data and user profiles
Static assets and computed results

⚖️ Trade-offs

Cache invalidation is hard
Stale data if TTL too long
Cache stampede on cold start
Memory is expensive vs disk

🎓 Interview Tips

Write-through vs write-back vs write-around
Cache stampede: probabilistic expiry, locks
L1/L2/L3 CPU cache = same concept
CDN is a cache at the network edge