[SERVER-1815] instrument more detailed response time items in profiler Created: 20/Sep/10 Updated: 12/Jul/16 Resolved: 08/Aug/11 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Admin, Performance |
| Affects Version/s: | None |
| Fix Version/s: | 1.9.2 |
| Type: | New Feature | Priority: | Major - P3 |
| Reporter: | Kenny Gorman | Assignee: | Eliot Horowitz (Inactive) |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||
| Participants: | |||||||||||||
| Description |
|
The profiler is awesome. But we need more data about the details of where the time went so we can tune individual components and so we can construct a call graph. Splitting the "Millis" component out by type would be an excellent first step. I realize it may be hard with MM files, but perhaps just treat all calls to MM files as I/O for now. We would not be able to see physical vs logical I/O, but ok, thats a compromise. But understanding I/O vs other items would be excellent! Especially if a write was waiting to get time on the writer process. I would love something like: read: , } write: I think the various types that makes sense are:
The idea of DB profiling is not new. The Oracle world as a great deal of expertise in this area, and it's one of it's great strengths. If you are not already familar here is a paper by Cary Milsap (who IMHO is the master of all things profiling in Oracle): |
| Comments |
| Comment by auto [ 08/Aug/11 ] |
|
Author: {u'login': u'erh', u'name': u'Eliot Horowitz', u'email': u'eliot@10gen.com'}Message: cleaning new profiler data |
| Comment by Eliot Horowitz (Inactive) [ 08/Aug/11 ] |
|
The profiler output has changed and there are more things now, but some are still missing |
| Comment by Kenny Gorman [ 25/Feb/11 ] |
|
Can you schedule this sooner? Like 1.8? |
| Comment by Eliot Horowitz (Inactive) [ 21/Sep/10 ] |
|
Some of these make sense - others would be harder. |
| Comment by Kenny Gorman [ 20/Sep/10 ] |
|
Oh, and another little data format tweak to make map reducing output easier: { "ts" : "Mon Sep 20 2010 11:55:52 GMT-0700 (PDT)", } |