5 collocation exercises on cache behaviour and management.
0 / 5 completed
1 / 5
Before peak traffic, you ___ the cache.
You warm a cache — pre-populating it with likely-requested data so early users get fast responses instead of cold misses. Warm is the established collocation (a warm cache versus a cold cache). Heat off, fire out and even prime up are not standard here. Cache warming is common before launches or after deploys that clear the cache, preventing a sudden surge of slow uncached requests.
2 / 5
When the source data changes, you ___ the cache.
You invalidate the cache — marking entries stale so the next request fetches fresh data. Invalidate is the precise collocation (invalidate a key, cache invalidation). Cancel off, void up and scrub out are wrong. As the famous saying goes, cache invalidation is one of the hardest problems in computing, because deciding exactly when cached data is no longer valid is genuinely tricky.
3 / 5
To control how long entries live, you ___ a TTL.
You set a TTL (time to live) — the expiry duration after which a cache entry is automatically discarded. The collocation is set a TTL (or set an expiry). Lay off, place up and fix out are not idiomatic. A sensible TTL balances freshness against load: short TTLs keep data current but increase backend hits, while long TTLs reduce load but risk serving stale data.
4 / 5
When the cache is full, it must ___ old entries.
The cache evicts entries — removing them to make room, typically using a policy like LRU (least recently used). Evict is the standard term (eviction policy). Expel off, oust up and boot out are not idiomatic in this context. Eviction differs from expiry: expiry is time-based via TTL, while eviction is triggered by capacity limits when the cache runs out of space.
5 / 5
A request found in cache is a ___; a missing one is a miss.
A cache hit means the requested data was found in the cache; the opposite is a cache miss. We talk about hit rate and say a request hits or misses the cache. Strike, touch and catch are not used. A high hit rate is the goal: more hits mean fewer expensive trips to the database or origin, lowering latency and load substantially.