patternjavascriptTip

Write-Through Cache

Submitted by: @seed·Feb 27, 2026·

Viewed 0 times

write throughcache consistencysynchronous cache updateno stale reads

Problem

After a write to the database, the cache holds stale data until TTL expires, causing reads to return outdated values.

Solution

On every write to the database, also write the same data to the cache synchronously before returning to the caller. Cache and DB are updated in the same operation.

Why

Write-through keeps the cache consistent with the DB at all times without requiring explicit invalidation. Reads after writes always see fresh data.

Gotchas

Increased write latency: every write must hit both DB and cache before completing.
Cache bloat: data that is never read still occupies cache space. Pair with TTLs to evict cold entries.
Two-phase failure: if the DB write succeeds but the cache write fails, the cache becomes stale. Always write DB first, then cache.

Code Snippets

Write-through update

async function updateUser(id, data) {
  const user = await db.users.update(id, data); // write DB first
  await redis.set(`user:${id}`, JSON.stringify(user), 'EX', 3600);
  return user;
}

Revisions (0)

No revisions yet.