* [ENHANCEMENT] TSDB: Improve calculation of space used by labels
The labels for each series in the Head take up some some space in the
Postings index, but far more space in the `memSeries` structure.
Instead of having the Postings index calculate this overhead, which is
a layering violation, have the caller pass in a function to do it.
Provide three implementations of this function for the three Labels
versions.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
While investigating lock contention on `MemPostings`, we saw that lots
of locking is happening in `LabelValues` and
`PostingsForLabelsMatching`, both copying the label values slices while
holding the mutex.
This adds an extra map that holds an append-only label values slice for
each one of the label names. Since the slice is append-only, it can be
copied without holding the mutex.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Same as #15427 but for the new method added in #14144
Instead of allocating each ListPostings one by one, allocate them all in
one go.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Instead of allocating ListPostings pointers one by one, allocate a slice
and take pointers from that. It's faster, and also generates less
garbage (NewListPostings is one of the top offenders in number of
allocations).
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Simple follow-up to #13620. Modify `tsdb.PostingsForMatchers` to use the optimized tsdb.IndexReader.PostingsForLabelMatching method also for inverse matching.
Introduce method `PostingsForAllLabelValues`, to avoid changing the existing method.
The performance is much improved for a subset of the cases; there are up to
~60% CPU gains and ~12.5% reduction in memory usage.
Remove `TestReader_InversePostingsForMatcherHonorsContextCancel` since
`inversePostingsForMatcher` only passes `ctx` to `IndexReader` implementations now.
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
This introduces back some unlocking that was removed in #13286 but in a
more balanced way, as suggested by @pracucci.
For TSDBs with a lot of churn, Delete() can take a couple of seconds,
and while it's holding the mutex, reads and writes are blocked waiting
for that mutex, increasing the number of connections handled and memory
usage.
This implementation pauses every 4K labels processed (note that also
compared to #13286 we're not processing all the label-values anymore,
but only the affected ones, because of #14307), makes sure that it's
possible to get the read lock, and waits for a few milliseconds more.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Co-authored-by: Marco Pracucci <marco@pracucci.com>
This reverts commit 50ef0dc954.
Memory allocation goes so high in Prombench that the system is unusable.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Tests for Mempostings.{Add,Get} data race
* Fix MemPostings.{Add,Get} data race
We can't modify the postings list that are held in MemPostings as they
might already be in use by some readers.
* Modify BenchmarkHeadStripeSeriesCreate to have common labels
If there are no common labels on the series, we don't excercise the
ordering part of MemSeries, as we're just creating slices of one element
for each label value.
---------
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
We are still seeing lock contention on MemPostings.mtx, and MemPostings.Delete() is by far the most expensive operation on that mutex.
This adds parallelism to that method, trying to reduce the amount of time we spend with the mutex held.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Go's built-in append() grows larger slices with factor 1.3, which means we do a lot more allocating and copying for larger postings.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Several things done here:
- Set `max-issues-per-linter` to 0 so that we actually see all linter
warnings and not just 50 per linter. (As we also set
`max-same-issues` to 0, I assume this was the intention from the
beginning.)
- Stop using the golangci-lint default excludes (by setting
`exclude-use-default: false`. Those are too generous and don't match
our style conventions. (I have re-added some of the excludes
explicitly in this commit. See below.)
- Re-add the `errcheck` exclusion we have used so far via the
defaults.
- Exclude the signature requirement `govet` has for `Seek` methods
because we use non-standard `Seek` methods a lot. (But we keep other
requirements, while the default excludes completely disabled the
check for common method segnatures.)
- Exclude warnings about missing doc comments on exported symbols. (We
used to be pretty adamant about doc comments, but stopped that at
some point in the past. By now, we have about 500 missing doc
comments. We may consider reintroducing this check, but that's
outside of the scope of this commit. The default excludes of
golangci-lint essentially ignore doc comments completely.)
- By stop using the default excludes, we now get warnings back on
malformed doc comments. That's the most impactful change in this
commit. It does not enforce doc comments (again), but _if_ there is
a doc comment, it has to have the recommended form. (Most of the
changes in this commit are fixing this form.)
- Improve wording/spelling of some comments in .golangci.yml, and
remove an outdated comment.
- Leave `package-comments` inactive, but add a TODO asking if we
should change that.
- Add a new sub-linter `comment-spacings` (and fix corresponding
comments), which avoids missing spaces after the leading `//`.
Signed-off-by: beorn7 <beorn@grafana.com>
* Pass affected labels to MemPostings.Delete
As suggested by @bboreham, we can track the labels of the deleted series
and avoid iterating through all the label/value combinations.
This looks much faster on the MemPostings.Delete call. We don't have a
benchmark on stripeSeries.gc() where we'll pay the price of iterating
the labels of each one of the deleted series.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
The only call we have to LabelValuesFor() has an index.Postings, and we
expand it to pass to this method, which will iterate over the values.
That's a waste of resources: we can iterate on the index.Postings
directly.
If there's any downstream implementation that has a slice of series,
they can always do an index.ListPostings from them: doing that is
cheaper than expanding an abstract index.Postings.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
* MemPostings.PostingsForLabelMatching: let mutex go
This changes the `MemPostings.PostingsForLabelMatching` implementation
to stop holding the read mutex while matching the label values.
We've seen that this method can be slow when the matcher is expensive,
that's why we even added a context expiration check.
However, there are critical process that might be waiting on this mutex:
writes (adding new series) and compaction (deleting the
garbage-collected ones), so we should avoid holding it for a long period
of time.
Given that we've copied the values to a slice anyway, there's no need to
hold the lock while matching.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
* MemPostings: reduce locking/unlocking
MemPostings.Delete is called from Head.gc(), i.e. it gets the IDs of the
series that have churned.
I'd assume that many label values aren't affected by that churn at all,
so it doesn't make sense to touch the lock while checking them.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Follow up on https://github.com/prometheus/prometheus/pull/14096
As promised, I bring a benchmark, which shows a very small improvement
if context is checked every 128 iterations of label instead of every
100.
It's much easier for a computer to check modulo 128 than modulo 100.
This is a very small 0-2% improvement but I'd say this is one of the
hottest paths of the app so this is still relevant.
Signed-off-by: Oleg Zaytsev <mail@olegzaytsev.com>
Followup to #14096
Unfortunately the previous PR introduced this bug by not releasing the
lock before returning.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* tsdb: check for context cancel before regex matching postings
Regex matching can be heavy if the regex takes a lot of cycles to
evaluate and we can get stuck evaluating postings for a long time
without this fix. The constant checkContextEveryNIterations=100
may be changed later.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Add method `PostingsForLabelMatching` to `tsdb.IndexReader`, to obtain postings for labels with a certain name and values accepted by a provided callback, and use it from `tsdb.PostingsForMatchers`.
The intention is to optimize regexp matcher paths, especially not having to load all label values before matching on them.
Plus tests, and refactor some `tsdb/index.Reader` methods.
Benchmarking shows memory reduction up to ~100%, and speedup of up to ~50%.
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
