VictoriaMetrics/lib/logstorage/stats_count_empty.go
Aliaksandr Valialkin 4599429f51
lib/logstorage: read timestamps column when it is really needed during query execution
Previously timestamps column was read unconditionally on every query.
This could significantly slow down queries, which do not need reading this column
like in https://github.com/VictoriaMetrics/VictoriaMetrics/issues/7070 .
2024-09-25 19:17:47 +02:00

207 lines
4.7 KiB
Go

package logstorage
import (
"slices"
"strconv"
"unsafe"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
)
type statsCountEmpty struct {
fields []string
}
func (sc *statsCountEmpty) String() string {
return "count_empty(" + statsFuncFieldsToString(sc.fields) + ")"
}
func (sc *statsCountEmpty) updateNeededFields(neededFields fieldsSet) {
updateNeededFieldsForStatsFunc(neededFields, sc.fields)
}
func (sc *statsCountEmpty) newStatsProcessor() (statsProcessor, int) {
scp := &statsCountEmptyProcessor{
sc: sc,
}
return scp, int(unsafe.Sizeof(*scp))
}
type statsCountEmptyProcessor struct {
sc *statsCountEmpty
rowsCount uint64
}
func (scp *statsCountEmptyProcessor) updateStatsForAllRows(br *blockResult) int {
fields := scp.sc.fields
if len(fields) == 0 {
bm := getBitmap(br.rowsLen)
bm.setBits()
for _, c := range br.getColumns() {
values := c.getValues(br)
bm.forEachSetBit(func(idx int) bool {
return values[idx] == ""
})
}
scp.rowsCount += uint64(bm.onesCount())
putBitmap(bm)
return 0
}
if len(fields) == 1 {
// Fast path for count_empty(single_column)
c := br.getColumnByName(fields[0])
if c.isConst {
if c.valuesEncoded[0] == "" {
scp.rowsCount += uint64(br.rowsLen)
}
return 0
}
if c.isTime {
return 0
}
switch c.valueType {
case valueTypeString:
for _, v := range c.getValuesEncoded(br) {
if v == "" {
scp.rowsCount++
}
}
return 0
case valueTypeDict:
zeroDictIdx := slices.Index(c.dictValues, "")
if zeroDictIdx < 0 {
return 0
}
for _, v := range c.getValuesEncoded(br) {
if int(v[0]) == zeroDictIdx {
scp.rowsCount++
}
}
return 0
case valueTypeUint8, valueTypeUint16, valueTypeUint32, valueTypeUint64, valueTypeFloat64, valueTypeIPv4, valueTypeTimestampISO8601:
return 0
default:
logger.Panicf("BUG: unknown valueType=%d", c.valueType)
return 0
}
}
// Slow path - count rows containing empty value for all the fields enumerated inside count_empty().
bm := getBitmap(br.rowsLen)
defer putBitmap(bm)
bm.setBits()
for _, f := range fields {
c := br.getColumnByName(f)
if c.isConst {
if c.valuesEncoded[0] != "" {
return 0
}
continue
}
if c.isTime {
return 0
}
switch c.valueType {
case valueTypeString:
valuesEncoded := c.getValuesEncoded(br)
bm.forEachSetBit(func(i int) bool {
return valuesEncoded[i] == ""
})
case valueTypeDict:
if !slices.Contains(c.dictValues, "") {
return 0
}
valuesEncoded := c.getValuesEncoded(br)
bm.forEachSetBit(func(i int) bool {
dictIdx := valuesEncoded[i][0]
return c.dictValues[dictIdx] == ""
})
case valueTypeUint8, valueTypeUint16, valueTypeUint32, valueTypeUint64, valueTypeFloat64, valueTypeIPv4, valueTypeTimestampISO8601:
return 0
default:
logger.Panicf("BUG: unknown valueType=%d", c.valueType)
return 0
}
}
scp.rowsCount += uint64(bm.onesCount())
return 0
}
func (scp *statsCountEmptyProcessor) updateStatsForRow(br *blockResult, rowIdx int) int {
fields := scp.sc.fields
if len(fields) == 0 {
for _, c := range br.getColumns() {
if v := c.getValueAtRow(br, rowIdx); v != "" {
return 0
}
}
scp.rowsCount++
return 0
}
if len(fields) == 1 {
// Fast path for count_empty(single_column)
c := br.getColumnByName(fields[0])
if c.isConst {
if c.valuesEncoded[0] == "" {
scp.rowsCount++
}
return 0
}
if c.isTime {
return 0
}
switch c.valueType {
case valueTypeString:
valuesEncoded := c.getValuesEncoded(br)
if v := valuesEncoded[rowIdx]; v == "" {
scp.rowsCount++
}
return 0
case valueTypeDict:
valuesEncoded := c.getValuesEncoded(br)
dictIdx := valuesEncoded[rowIdx][0]
if v := c.dictValues[dictIdx]; v == "" {
scp.rowsCount++
}
return 0
case valueTypeUint8, valueTypeUint16, valueTypeUint32, valueTypeUint64, valueTypeFloat64, valueTypeIPv4, valueTypeTimestampISO8601:
return 0
default:
logger.Panicf("BUG: unknown valueType=%d", c.valueType)
return 0
}
}
// Slow path - count the row at rowIdx if at least a single field enumerated inside count() is non-empty
for _, f := range fields {
c := br.getColumnByName(f)
if v := c.getValueAtRow(br, rowIdx); v != "" {
return 0
}
}
scp.rowsCount++
return 0
}
func (scp *statsCountEmptyProcessor) mergeState(sfp statsProcessor) {
src := sfp.(*statsCountEmptyProcessor)
scp.rowsCount += src.rowsCount
}
func (scp *statsCountEmptyProcessor) finalizeStats() string {
return strconv.FormatUint(scp.rowsCount, 10)
}
func parseStatsCountEmpty(lex *lexer) (*statsCountEmpty, error) {
fields, err := parseStatsFuncFields(lex, "count_empty")
if err != nil {
return nil, err
}
sc := &statsCountEmpty{
fields: fields,
}
return sc, nil
}