Redis Caching Strategies in Code

This lesson has a runnable companion file at src/exercises/redis-caching-strategies.ts. Open it, read through each strategy, then run it against a local Redis instance to see the behavior.

npx tsx src/exercises/redis-caching-strategies.ts

What the Exercise Covers

The RedisCacheStrategy class implements five patterns:

1. Cache-Aside (Lazy Loading)

The application checks the cache, and on a miss it fetches from the database and populates the cache itself.

await cache.cacheAsideGet("user:1");  // MISS → fetches DB → fills cache
await cache.cacheAsideGet("user:1");  // HIT

2. Read-Through

Same result as cache-aside, but the cache layer is responsible for calling the database via a loader callback. The calling code never touches the DB.

await cache.readThrough("product:42", dbGet);  // cache calls loader on miss

3. Write-Through

Every write goes to the cache and the database synchronously. The cache is always consistent, but writes are slower.

await cache.writeThrough("order:99", "confirmed");
// Both cache and DB are updated before this line runs

4. Write-Behind (Write-Back)

Writes go to the cache immediately. A background interval flushes queued writes to the database in batches.

cache.startWriteBehindFlush(1000);
await cache.writeBehind("metric:views", "12345");
// Cache is updated instantly; DB catches up after ~1 second

5. Refresh-Ahead

Proactively refreshes a key in the background when its TTL drops below a threshold, keeping hot data warm.

await cache.refreshAhead("session:abc", dbGet, 10);
// If TTL < 10s, triggers a non-blocking background reload

Strategy Comparison

Strategy	Read Latency	Write Latency	Consistency	Data Loss Risk
Cache-Aside	Miss penalty	Low (DB only)	Eventual	None
Read-Through	Miss penalty	N/A	Eventual	None
Write-Through	Always fast	High (both)	Strong	None
Write-Behind	Always fast	Low (cache only)	Eventual	If cache crashes
Refresh-Ahead	Always fast	N/A	Eventual	None

Things to Try

Change the defaultTTL to a low value (e.g. 5 seconds) and observe keys expiring
Add console.log statements inside dbGet / dbSet to see when the database is actually hit
Break the write-behind flush and observe that the database never gets updated
Combine read-through with write-behind — read via readThrough, write via writeBehind