feat(bloomstore): Introduce specialised cache for blocks #12257
Merged
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
chaudum
force-pushed
the
chaudum/blocks-cache
branch
2 times, most recently
from
2771174
to
b99d663
Compare
The blocks cache is an in-memory cache that represents the single source of truth of what blocks are available on the filesystem and manages their EOL lifecycle. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
chaudum
force-pushed
the
chaudum/blocks-cache
branch
from
b99d663
to
d66c174
Compare
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
|
Regarding commit a1ad00d Before: After: |
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
salvacorts
reviewed
Mar 20, 2024
| _, found = cache.Get(ctx, "b") | ||
| require.False(t, found) | ||
|
|
||
| require.Equal(t, int64(8), cache.currSizeBytes) |
There was a problem hiding this comment.
I'd add here another key which is older than b but has a refcount and therefore shouldn't be evicted.
There was a problem hiding this comment.
there is actually a bug in the LRU eviction job
|
|
||
| // write | ||
| for k, v := range entries { | ||
| err := cache.Store(ctx, []string{k}, []BlockDirectory{v}) |
There was a problem hiding this comment.
nit: IIUC this is the cache that we want to replace with the new implementation. Maybe we can rename the new PutMany by Store so we don't need to change it wherever we use it.
There was a problem hiding this comment.
In the bloomstore we use only ever use Put() to add a single entry.
In the benchmark I wanted to make the calls the same, though.
| require.NoError(t, err) | ||
|
|
||
| err = cache.Put(ctx, "key", CacheValue("b", 10)) | ||
| require.ErrorContains(t, err, "entry already exists: key") |
There was a problem hiding this comment.
nit: maybe we can use errAlreadyExists right away so if one changes, the test does as well.
|
|
||
| ctx := context.Background() | ||
|
|
||
| _ = cache.Put(ctx, "a", CacheValue("a", 5)) |
There was a problem hiding this comment.
I think we should test PutInc here.
| return nil | ||
| } | ||
|
|
||
| // NewFsBlocksCache returns a new initialised BlocksCache with the key and value of requested types. |
There was a problem hiding this comment.
I think this comment is outdated
| return entry.Value, true | ||
| } | ||
|
|
||
| func (c *BlocksCache) get(_ context.Context, key string) *Entry { |
There was a problem hiding this comment.
if ctx is unused, I think we should remove it.
| close(c.done) | ||
| } | ||
|
|
||
| func (c *BlocksCache) sizeOf(entry *Entry) int64 { |
There was a problem hiding this comment.
nit: I' rather use entry.Value.Size() right away instead of this method.
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
salvacorts
approved these changes
Mar 20, 2024
Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
edsoncelio
pushed a commit
to edsoncelio/loki
that referenced
this pull request
Mar 22, 2024
The blocks cache is an in-memory cache that represents the single source of truth of what blocks are available on the file system and manages their end-of-life behaviour. There is a follow-up branch to wire up the new cache, based on this branch. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
rhnasc
pushed a commit
to inloco/loki
that referenced
this pull request
Apr 12, 2024
The blocks cache is an in-memory cache that represents the single source of truth of what blocks are available on the file system and manages their end-of-life behaviour. There is a follow-up branch to wire up the new cache, based on this branch. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>
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.
What this PR does / why we need it:
The blocks cache is an in-memory cache that represents the single source of truth of what blocks are available on the file system and manages their EOL lifecycle.
Special notes for your reviewer:
There is a follow-up branch to wire up the new cache, based on this branch. See https://github.com/grafana/loki/compare/chaudum/blocks-cache-impl?expand=1