-
Type:
Improvement
-
Resolution: Won't Fix
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: Admin, Performance, Storage, Usability
-
Environment:all
This is important to be able to "hint" to the system the correct padding, as some workloads are heavy to start and tapper off. This type of workload makes it impossible to learn the padding factor when needed. It would not override the normal behavior just tune the current level for heuristics to still tune as more data is learned.
It should behave like this, so that is not a new setting up just updating the existing data to help the engine make better INITIAL decisions.
> db.col1.stats()
{
"ns" : "foo.col1",
"count" : 1,
"size" : 24,
"avgObjSize" : 24,
"storageSize" : 4096,
"numExtents" : 1,
"nindexes" : 1,
"lastExtentSize" : 4096,
"paddingFactor" : 1,
"systemFlags" : 1,
"userFlags" : 0,
"totalIndexSize" : 8176,
"indexSizes" : {
"_id_" : 8176
},
"ok" : 1
}
>db.runCommand( {"collMod" : "col1" , "paddingFactor" : 1.3 } )
{
"PaddingSize_old" : 1,
"PaddingSize_new" : 1.3,
"ok" : 1
}
> db.col1.stats()
{
"ns" : "foo.col1",
"count" : 1,
"size" : 24,
"avgObjSize" : 24,
"storageSize" : 4096,
"numExtents" : 1,
"nindexes" : 1,
"lastExtentSize" : 4096,
"paddingFactor" : 1.3,
"systemFlags" : 1,
"userFlags" : 0,
"totalIndexSize" : 8176,
"indexSizes" : {
"_id_" : 8176
},
"ok" : 1
}
Pull request is ready once the test case is written.
- is related to
-
-
- Closed
-
- links to
