This reduction is based on production testing.
Also expose -search.minWindowForInstantRollupOptimization command-line flag, so users could fine-tune this arg for their needs
Repeated instant queries with long lookbehind windows, which contain one of the following rollup functions,
are optimized via partial result caching:
- sum_over_time()
- count_over_time()
- avg_over_time()
- increase()
- rate()
The basic idea of optimization is to calculate
rf(m[d] @ t)
as
rf(m[offset] @ t) + rf(m[d] @ (t-offset)) - rf(m[offset] @ (t-d))
where rf(m[d] @ (t-offset)) is cached query result, which was calculated previously
The offset may be in the range of up to 1 hour.
The new metric gets increased each time `-search.logQueryMemoryUsage` memory limit
is exceeded by a query. This metric should help to identify expensive and heavy queries
without inspecting the logs.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
This can be useful in the following queries:
drop_empty_series(temperature <= 30) default 40
This query drops temperature series with all the values bigger than 30 on the selected time range,
while replacing gaps in the remaining series with 40.
The query without drop_empty_series:
(temperature <= 30) default 40
would leave all the temperature series with all the values bigger than 30 on the selected time range,
and replace all their values with 40. This is not what could be epxected in some cases
like here - https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5071
reduce lock contention for heavy aggregation requests
previously lock contetion may happen on machine with big number of CPU due to enabled string interning. sync.Map was a choke point for all aggregation requests.
Now instead of interning, new string is created. It may increase CPU and memory usage for some cases.
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5087
`median_over_time` is handled by predefined WITH template in MetricsQL library which translates it to `quantile_over_time(0.5)`
This makes it impossble to use `median_over_time` as a usual rollup function for `aggr_over_time`.
See: https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5034
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
This reverts commit 252643d100.
Reason for revert: the commit incorrectly fixes the the issue.
The `remoteAddr` must be properly quoted inside lib/httpserver.GetQuotedRemoteAddr().
It isn't quoted properly if the request contains X-Forwarded-For header.
The proper fix will be included in the follow-up commit.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/4676
This eliminates the need in .(*T) casting for results obtained from Load()
Leave atomic.Value for map, since atomic.Pointer[map[...]...] makes double pointer to map,
because map is already a pointer type.
- Add `Active queries` chapter to VMUI docs
- Set `Content-Type: json` header inside promql.WriteActiveQueries() handler,
in order to be consistent with other request handlers called at app/vmselect/main.go
- Pass the request to promql.WriteActiveQueries() handler, so it can change its output
depending on the provided request params. This also improves consistency of
promql.WriteActiveQueries() args with other request hanlers at app/vmselect/main.go
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/4653
* feat: add page to display a list of active queries (#4598)
* app/vmagent: code formatting
* fix: remove console
---------
Co-authored-by: dmitryk-dk <kozlovdmitriyy@gmail.com>
The `a op b keep_metric_names` is ambigouos to `a op (b keep_metric_names)` when `b` is a transform or rollup function.
For example, `a + rate(b) keep_metric_names`. So it is better to use more clear syntax: `(a op b) keep_metric_names`
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3710
* metricsql: add support of using keep_metric_names for binary operations
This should help to avoid confusion with queries like one in the issue #3710.
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
* wip
---------
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
Callers of this function log the returned error and exit.
So let's just log the error with the given filepath and the call stack
inside the function itself and then exit. This simplifies the code
at callers' place while leaves the same level of debuggability in case of errors.
There is a bug here where if you have a single bucket like:
foo{vmrange="4.084e+02...4.642e+02"} 2 123
The expected output is three le encoded buckets like:
foo{le="4.084e+02"} 0 123
foo{le="4.642e+02"} 2 123
foo{le="+Inf"} 2 123
This correctly encodes the start and end of the vmrange.
If however, the input contains the previous bucket, and that bucket is
empty then you only get the end le and +Inf out currently, i.e:
foo{vmrange="7.743e+05...8.799e+05"} 5 123
foo{vmrange="6.813e+05...7.743e+05"} 0 123
results in:
foo{le="8.799e+05"} 5 123
foo{le="+Inf"} 5 123
This causes issues when you go to compute a quantile because this means
that the assumed lower bound of the buckets is 0 and this we interpolate
between 0->end rather than the vmrange start->end as expected.
- Expose stats.seriesFetched at `/api/v1/query_range` responses too
for the sake of consistency.
- Initialize QueryStats when it is needed and pass it to EvalConfig then.
This guarantees that the QueryStats is properly collected when the query
contains some subqueries.
The change adds a new field `seriesFetched` to EvalConfig object.
Since EvalConfig object can be copied inside `Exec`,
`seriesFetched` is a pointer which can be updated by all copied
objects.
The reason for having stats is that other components, like vmalert,
could benefit from this information.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
- Allocate and initialize seriesByWorkerID slice in a single go instead
of initializing every item in the list separately.
This should reduce CPU usage a bit.
- Properly set anti-false sharing padding at timeseriesWithPadding structure
- Document the change at docs/CHANGELOG.md
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3966
* vmselect/promql: refactor `evalRollupNoIncrementalAggregate` to use lock-less approach for parallel workers computation
Locking there is causing issues when running on highly multi-core system as it introduces lock contention during results merge.
New implementation uses lock less approach to store results per workerID and merges final result in the end, this is expected to significantly reduce lock contention and CPU usage for systems with high number of cores.
Related: #3966
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
* vmselect/promql: add pooling for `timeseriesWithPadding` to reduce allocations
Related: #3966
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
* vmselect/promql: refactor `evalRollupFuncWithSubquery` to avoid using locks
Uses same approach as `evalRollupNoIncrementalAggregate` to remove locking between workers and reduce lock contention.
Related: #3966
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
---------
Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
This opens the possibility to remove tssLock from evalRollupFuncWithSubquery()
in the follow-up commit from @zekker6 in order to speed up the code
for systems with many CPU cores.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3966
* vmselect/promql: check for deadline in `count_values` fn
`count_values` could be very slow during the data processing.
Checking for deadline between iterations supposed to reduce
probability of exceeding `search.maxQueryDuration`.
The change also adds a new trace record, which captures the time
spent in aggregation function. Before that, the trace for aggr funcs
could be confusing since it doesn't account for all the places where
time was spent.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
* wip
---------
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
* metricsql: support optional 2nd argument for rollup functions
Support optional 2nd argument `min`, `max` or `avg` for rollup functions:
* rollup
* rollup_delta
* rollup_deriv
* rollup_increase
* rollup_rate
* rollup_scrape_interval
If second argument is passed, then rollup function will return only the selected aggregation type.
This change can be useful for situations where only one type of rollup calculation is needed.
For example, `rollup_rate(requests_total[5m], "max")`.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
* wip
---------
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
Previously too short lookbehind window d for rate(m[d]) could be automatically extended
if it didn't cover at least two raw samples. This was needed in order to guarantee
non-empty results from rate(m[d]) on short time ranges.
Now the lookbehind window isn't extended if it is set explicitly,
since it is expected that the user knows what he is doing.
The lookbehind window continues to be extended when needed if it isn't set explicitly.
For example, in the case of rate(m).
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3483
Previously empty series (e.g. series with all NaN samples) were passed to aggregate functions.
Such series must be ingored by all the aggregate functions.
So it is better from consistency PoV filtering out empty series before applying aggregate functions.
* app/vmselect: ignore empty series for `limit_offset`
VictoriaMetrics doesn't return empty series (with all NaN values) to
the user. But such series are filtered after transform functions.
It means `limit_offset` will account for empty series as well.
For example, let's consider following data set:
```
time series:
foo{label="1"} NaN, NaN, NaN, NaN // empty series
foo{label="2"} 1, 2, 3, 4
foo{label="3"} 4, 3, 2, 1
```
When user requests all series for metric `foo` the empty series
will be filtered out:
```
/query=foo:
foo{label="v2"} 1, 2, 3, 4
foo{label="v3"} 4, 3, 2, 1
```
But `limit_offset(1, 1, foo)` is applied to original series, not filtered yet.
So it will return `foo{label="v2"}` (skips the first in list)
```
/query=limit_offset(1, 1, foo):
foo{label="v2"} 1, 2, 3, 4
```
Expected result would be to apply `limit_offset` to already filtered list,
so in result we receive `foo{label="v3"}`:
```
/query=limit_offset(1, 1, foo):
foo{label="v3"} 4, 3, 2, 1
```
The change does exactly that - filters empty series before applying `limit_offset`.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
* app/vmselect: ignore empty series for `limit_offset`
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Signed-off-by: hagen1778 <roman@victoriametrics.com>
The workaround was introduced to fix https://github.com/VictoriaMetrics/VictoriaMetrics/issues/962.
However, it didn't prove itself useful. Instead, it is recommended using `increase_pure` function.
Removing the workaround makes VM to produce accurate results when calculating
`delta` or `increase` functions over slow-changing counters with vary intervals
between data points.
Signed-off-by: hagen1778 <roman@victoriametrics.com>
Signed-off-by: hagen1778 <roman@victoriametrics.com>
