[SERVER-84049] Deprecate support for sharding time-series collections on time field Created: 11/Dec/23 Updated: 14/Dec/23 |
|
| Status: | Blocked |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Henrik Edin | Assignee: | Backlog - Storage Execution Team |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||
| Assigned Teams: |
Storage Execution
|
||||
| Participants: | |||||
| Description |
|
Allowing time-series collections to be sharded on the time field adds complexity and special handling for time-series in code. Because a time-series bucket contains a range of time and may overlap it is impossible without certain restrictions (such as fixed bucketing) to determine which shard owns a particular time. It makes building features such as unique indexes for time-series collections more difficult and inefficient. Users with large time-series workloads typically have (very) high cardinality and sharding on meta is sufficient. By deprecating support for sharding time-series collections on the time field, we can remove it in the future to allow for simplifications of the code. |