Improve tsm1 cache performance #7228
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
stevenh
force-pushed
the
cache-lock-contention
branch
from
5bcecb9
to
97d7015
Compare
Reduce the cache lock contention by widening the cache lock scope in WriteMulti, while this sounds counter intuitive it was: * 1 x Read Lock to read the size * 1 x Read Lock per values * 1 x Write Lock per values on race * 1 x Write Lock to update the size We now have: * 1 x Write Lock This also reduces contention on the entries Values lock too as we have the global cache lock. Move the calculation of the added size before taking the lock as it takes time and doesn't need the lock. This also fixes a race in WriteMulti due to the lock not being held across the entire operation, which could cause the cache size to have an invalid value if Snapshot has been run in the between the addition of the values and the size update. Fix the cache benchmark which where benchmarking the creation of the cache not its operation and add a parallel test for more real world scenario, however this could still be improved. Add a fast path newEntryValues values for the new case which avoids taking the values lock and all the other calculations. Drop the lock before performing the sort in Cache.Keys().
stevenh
force-pushed
the
cache-lock-contention
branch
from
97d7015
to
ab61ed0
Compare
|
Rebased to eliminate conflict. |
rw
reviewed
Sep 23, 2016
| newSize := c.size + uint64(totalSz) | ||
| if c.maxSize > 0 && newSize+c.snapshotSize > c.maxSize { | ||
| c.mu.RUnlock() | ||
| c.mu.Lock() |
There was a problem hiding this comment.
Why did you switch this from a read-only lock to a read-write lock?
There was a problem hiding this comment.
In general we want to use RLock whenever possible.
There was a problem hiding this comment.
Because taking the lock multiple times was causing significant slowdown due to lock contention, but also as how this was structured was racey resulting in the potential for c.size to become totally invalid.
|
|
||
| var a []string | ||
| for k, _ := range c.store { | ||
| a = append(a, k) | ||
| } | ||
| c.mu.RUnlock() |
There was a problem hiding this comment.
Eliminating defers is always good :-)
|
@stevenh This is working well in my tests. Can you rebase? |
6 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Reduce the cache lock contention by widening the cache lock scope in WriteMulti, while this sounds counter intuitive it was:
We now have:
This also reduces contention on the entries Values lock too as we have the global cache lock.
Move the calculation of the added size before taking the lock as it takes time and doesn't need the lock.
This also fixes a race in WriteMulti due to the lock not being held across the entire operation, which could cause the cache size to have an invalid value if Snapshot has been run in between the addition of the values and the size update.
Fix the cache benchmark which where benchmarking the creation of the cache not its operation and add a parallel test for more real world scenario, however this could still be improved.
Add a fast path newEntryValues values for the new case which avoids taking the values lock and all the other calculations.