Architecting for Physical Partition Splitting Latency in High-Throughput Write Workloads
Updated Apr 28, 2026
Short answer
Partition splitting occurs when a physical partition hits 50GB or its RU limit. Architecture must account for the 50/50 throughput split to avoid temporary 'hot' child partitions.
Deep explanation
In Azure Cosmos DB, physical partitions are the units of scalability. When a partition reaches its storage (50GB) or throughput limit, it splits into two child partitions. The critical architectural nuance is that the provisioned RUs are split exactly 50/50 between the new partitions. If your data distribution is skewed, one child partition may immediately become a 'hot' partition, leading to 429 Throttle errors even if the total container RUs are under-utilized. High-throughput systems must use synthetic keys or hierarchical partitioning to ensure even distribution post-split.
Unlock with a Pro subscription to view this section.
View pricingReal-world example
No real-world example available yet.
Unlock with a Pro subscription to view this section.
Upgrade to ProCommon mistakes
No common mistakes listed yet.
Unlock with a Pro subscription to view this section.
Upgrade to ProFollow-up questions
No follow-up questions available yet.
Unlock with a Pro subscription to view this section.
Upgrade to Pro