DynamoDB Partition Key Design Language Collocations
Practise the standard verbs for designing DynamoDB partition keys that avoid hot partitions.
0 / 5 completed
1 / 5
Fill in: 'We ___ a partition key with high cardinality across every access pattern so a single hot partition doesn't throttle every request hitting one physical shard.'
We 'choose a partition key' — the standard, simple collocation for selecting the attribute that determines data distribution. The other options aren't idiomatic here.
2 / 5
Fill in: 'Picking a low-cardinality partition key like a status flag can ___ almost every write landing on the same physical partition, capping throughput regardless of table scale.'
We say a poor key choice will 'leave' writes concentrated on one partition — the standard, natural collocation for the resulting bottleneck. The other options aren't idiomatic here.
3 / 5
Fill in: 'We ___ per-partition consumed capacity continuously, since a single overloaded partition can throttle requests long before the table's aggregate capacity is exhausted.'
We 'monitor' capacity — the standard collocation for ongoing observation of a per-partition metric. The other options aren't idiomatic here.
4 / 5
Fill in: 'We ___ writes across a randomized key suffix so a naturally skewed access pattern doesn't concentrate every request onto one shard.'
We 'spread writes' — the standard, established collocation for distributing traffic across partitions using a suffix technique. The other options aren't the recognised term here.
5 / 5
Fill in: 'We ___ the partition key design against real production access patterns before launch, rather than trusting an assumption made early in the schema's design.'
We 'review' a design — the standard, simple collocation for validating a choice against real usage. The other options are less idiomatic here.