[SERVER-17502] Better bufbuilder size limit strategy for mmap journal Created: 06/Mar/15 Updated: 06/Dec/22 Resolved: 14/Sep/18 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | MMAPv1, Storage |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Alexander Komyagin | Assignee: | Backlog - Storage Execution Team |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||
| Assigned Teams: |
Storage Execution
|
||||
| Participants: | |||||
| Description |
|
Instead of crashing when the bufbuilder size reaches some hardcoded max-size, we can make it configurable (and call it "journal cache size"), and do a stop-the-world (pause all writes to it) if the max-size is reached. This strategy is better than what we have right now, as it allows the user to make a choice based on his concurrency requirements, in a somewhat similar fashion he can configure the cache size in WT. In the event of reaching max-size, the user won't see his server crashing but will experience a short performance degradation, which is better in most cases. |