[SERVER-7831] m/r: reduce post processing: reduce lock time with "nonAtomic" option Created: 04/Dec/12 Updated: 06/Dec/22 |
|
| Status: | Backlog |
| Project: | Core Server |
| Component/s: | MapReduce |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Azat Khuzhin | Assignee: | Backlog - Query Execution |
| Resolution: | Unresolved | Votes: | 3 |
| Labels: | pull-request | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
||||||||||||||||
| Issue Links: |
|
||||||||||||||||
| Assigned Teams: |
Query Execution
|
||||||||||||||||
| Participants: | |||||||||||||||||
| Description |
|
Merge too many documents, can take a long period of time, " option. This patch do this. |
| Comments |
| Comment by Benety Goh [ 10/Jul/14 ] |
|
Attached file containing pull request changes rebased on 81ed23939d7b94a14371241db247a76e85f050fe |
| Comment by Benety Goh [ 10/Jul/14 ] |
|
|
| Comment by Brent Miller [ 07/Jun/13 ] |
|
We think we've been affected by this issue, and just wanted to comment on how this effects performance. When the server enters the "M/R Reduce Post Processing" stage, the global lock will shoot to 100%, rendering the primary almost completely useless – queries time out, replica secondaries will fall behind, and the insert rate will drop dramatically. This more or less makes the out: {reduce: "collName"}map reduce option impractical to use in a production environment. I asked about this on the mongodb-users group not too long ago. If it provides any useful background information, the post can be found here. |
| Comment by Azat Khuzhin [ 10/Mar/13 ] |
|
Any news? |
| Comment by Azat Khuzhin [ 04/Dec/12 ] |
|
Pull request - https://github.com/mongodb/mongo/pull/333 |