Commit graph

12 commits

Author SHA1 Message Date
Zakhar Bessarab
9f9cc24e4c
Revert "lib/mergeset: add sparse indexdb cache (#7269)"
This reverts commit 837d0d136d.
2024-11-04 10:29:14 -03:00
Zakhar Bessarab
837d0d136d
lib/mergeset: add sparse indexdb cache (#7269)
Related issue:
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/7182

- add a separate index cache for searches which might read through large
amounts of random entries. Primary use-case for this is retention and
downsampling filters, when applying filters background merge needs to
fetch large amount of random entries which pollutes an index cache.
Using different caches allows to reduce effect on memory usage and cache
efficiency of the main cache while still having high cache hit rate. A
separate cache size is 5% of allowed memory.

- reduce size of indexdb/dataBlocks cache in order to free memory for
new sparse cache. Reduced size by 5% and moved this to a separate cache.

- add a separate metricName search which does not cache metric names -
this is needed in order to allow disabling metric name caching when
applying downsampling/retention filters. Applying filters during
background merge accesses random entries, this fills up cache and does
not provide an actual improvement due to random access nature.


Merge performance and memory usage stats before and after the change:

- before

![image](https://github.com/user-attachments/assets/485fffbb-c225-47ae-b5c5-bc8a7c57b36e)


- after

![image](https://github.com/user-attachments/assets/f4ba3440-7c1c-4ec1-bc54-4d2ab431eef5)

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
2024-10-24 15:21:17 +02:00
Aliaksandr Valialkin
0f24078146
lib/logstorage: use simpler in-memory cache instead of workingsetcache for caching recently ingested _stream values and recently queried set of streams
These caches aren't expected to grow big, so it is OK to use the most simplest cache based on sync.Map.
The benefit of this cache compared to workingsetcache is better scalability on systems with many CPU cores,
since it doesn't use mutexes at fast path.
An additional benefit is lower memory usage on average, since the size of in-memory cache equals
working set for the last 3 minutes.

The downside is that there is no upper bound for the cache size, so it may grow big during workload spikes.
But this is very unlikely for typical workloads.
2024-10-18 02:22:43 +02:00
Aliaksandr Valialkin
ad505a7a9a
lib/logstorage: work-in-progress 2024-05-20 04:08:30 +02:00
Aliaksandr Valialkin
cc2647d212
lib/encoding: optimize UnmarshalVarUint64, UnmarshalVarInt64 and UnmarshalBytes a bit
Change the return values for these functions - now they return the unmarshaled result plus
the size of the unmarshaled result in bytes, so the caller could re-slice the src for further unmarshaling.

This improves performance of these functions in hot loops of VictoriaLogs a bit.
2024-05-14 01:23:54 +02:00
Aliaksandr Valialkin
9dbd0f9085
lib/logstorage: initial implementation of pipes in LogsQL
See https://docs.victoriametrics.com/victorialogs/logsql/#pipes
2024-05-12 16:33:31 +02:00
Aliaksandr Valialkin
590160ddbb
lib/slicesutil: add helper functions for setting slice length and extending its capacity
The added helper functions - SetLength() and ExtendCapacity() - replace error-prone code with simple function calls.
2024-05-12 11:32:17 +02:00
Aliaksandr Valialkin
4617dc8bbe
lib/logstorage: consistently use atomic.* types instead of atomic.* functions on regular types
See ea9e2b19a5
2024-02-23 23:46:13 +02:00
Aliaksandr Valialkin
f81b480905
lib/mergeset: consistently use atomic.* types instead of atomic.* function calls on ordinary types
See ea9e2b19a5
2024-02-23 23:29:35 +02:00
Aliaksandr Valialkin
42dd71bb63
all: consistently use %w instead of %s in when error is passed to fmt.Errorf()
This allows consistently using errors.Is() for verifying whether the given error wraps some other known error.
2023-10-25 21:24:03 +02:00
Aliaksandr Valialkin
f35d27aa2b
app/vlstorage: expose vl_data_size_bytes metric at /metrics page for tracking the on-disk data size (both indexdb and the data itself) 2023-07-31 07:56:53 -07:00
Aliaksandr Valialkin
87b66db47d
app/victoria-logs: initial code release 2023-06-19 22:55:12 -07:00