-
Type: Task
-
Resolution: Done
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: Concurrency
-
None
A unified scheduling module would consolidate logic around what operations get dispatched in parallel with other operations (fsync+lock and reads, e.g.), could ultimately manage resources (limit iops per individual user op, or disk space dedicated to external sorts at once), and more.
TODO: Fill out the description w/ a fuller design proposal.
- has to be done before
-
SERVER-1423 reads often aren't possible while in fsync and lock mode
- Closed
- is related to
-
SERVER-3531 map reduce doesn't seem to yield unless it finds matches
- Closed
-
SERVER-5724 dbtemprelease's destructor can fail to restore state and assert on failure
- Closed
-
SERVER-8161 locks leak
- Closed
-
SERVER-8378 indexbg1.js failures
- Closed
-
SERVER-9147 touch command does not yield
- Closed
-
SERVER-3931 Shell hangs on tab complete if db is fsync&locked
- Closed
-
SERVER-8342 Delete old query_yield tests
- Closed
-
SERVER-1256 low priority write flag
- Closed
-
SERVER-1423 reads often aren't possible while in fsync and lock mode
- Closed
-
SERVER-3642 make sure yieldSometimes RecordNeeds specs are appropriate
- Closed
-
SERVER-6183 count may not yield before paging in initial candidate documents
- Closed
-
SERVER-8654 yielding too much?
- Closed
-
SERVER-3801 Mapreduce should yield periodically when writing
- Closed
-
SERVER-8307 Use db-level lock when MapReduce input and output use a single DB
- Closed
-
SERVER-3456 Running fsync+lock on config server during backup results in blocked inserts
- Closed
- related to
-
SERVER-11670 attempt to create a large collection in local db causes continuous asserts
- Closed