Improve *deduplicatingSlice.ingest performance (#4037)

* Add benchmark for `*PartitionWriter.WriteProfileSymbols`

I had to perform the initialization within the for loop as otherwise
it always panics.

* Refactor `*PartitionWriter.WriteProfileSymbols` for DRY

Thank you @simonswine for the suggestion!

Co-authored-by: Christian Simon <[email protected]>

* Create rewriting map with maximum capacity

By doing this we avoid having to regrow the map, which improves CPU
and memory consumption.

* Use temp slice to append only once

Instead of checking `s.slice` on each call to append and regrow if
necessary, create a temporary slice with the maximum possible
capacity, and append that slice to `s.slice` outside the for loop,
effectively checking and regrowing `s.slice` only once.

This further improves CPU and memory consumption.

* Use stdlib slices.Grow to grow the slice

---------

Co-authored-by: Christian Simon <[email protected]>

Author: inkel

pkg/phlaredb/symdb/dedup_slice.go

@@ -4,6 +4,7 @@ package symdb
 import (
 	"fmt"
 	"hash/maphash"
+	stdslices "slices"
 	"sort"
 	"sync"
 	"unsafe"
@@ -254,7 +255,7 @@ func (s *deduplicatingSlice[M, K, H]) Size() uint64 {
 
 func (s *deduplicatingSlice[M, K, H]) ingest(elems []M, rewriter *rewriter) {
 	var (
-		rewritingMap = make(map[int64]int64)
+		rewritingMap = make(map[int64]int64, len(elems))
 		missing      = int64SlicePool.Get()
 	)
 	missing = missing[:0]
@@ -279,6 +280,7 @@ func (s *deduplicatingSlice[M, K, H]) ingest(elems []M, rewriter *rewriter) {
 	if len(missing) > 0 {
 		s.lock.Lock()
 		posSlice := int64(len(s.slice))
+		s.slice = stdslices.Grow(s.slice, len(missing))
 		for _, pos := range missing {
 			// check again if element exists
 			k := s.helper.key(elems[pos])

pkg/phlaredb/symdb/symdb_test.go

@@ -227,3 +227,23 @@ func TestWritePartition(t *testing.T) {
 `
 	require.Equal(t, expected, resolved.String())
 }
+
+func BenchmarkPartitionWriter_WriteProfileSymbols(b *testing.B) {
+	b.ReportAllocs()
+
+	p, err := pprof.OpenFile("testdata/profile.pb.gz")
+	require.NoError(b, err)
+	p.Normalize()
+	cfg := DefaultConfig().WithDirectory(b.TempDir())
+
+	db := NewSymDB(cfg)
+
+	for i := 0; i < b.N; i++ {
+		b.StopTimer()
+		newP := p.CloneVT()
+		pw := db.PartitionWriter(uint64(i))
+		b.StartTimer()
+
+		pw.WriteProfileSymbols(newP)
+	}
+}