[#1328] pilorama: Do not skip items in SortedByFilename
All checks were successful
Tests and linters / Run gofumpt (pull_request) Successful in 2m46s
DCO action / DCO (pull_request) Successful in 3m13s
Vulncheck / Vulncheck (pull_request) Successful in 3m49s
Build / Build Components (1.22) (pull_request) Successful in 4m11s
Build / Build Components (1.23) (pull_request) Successful in 4m11s
Pre-commit hooks / Pre-commit (pull_request) Successful in 4m43s
Tests and linters / Tests (1.22) (pull_request) Successful in 4m49s
Tests and linters / Tests (1.23) (pull_request) Successful in 4m47s
Tests and linters / Staticcheck (pull_request) Successful in 4m47s
Tests and linters / Lint (pull_request) Successful in 5m8s
Tests and linters / gopls check (pull_request) Successful in 5m7s
Tests and linters / Tests with -race (pull_request) Successful in 5m13s

Benchmark results:
```
goos: linux
goarch: amd64
pkg: git.frostfs.info/TrueCloudLab/frostfs-node/pkg/local_object_storage/pilorama
cpu: 11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz
                                   │     old     │                 new                 │
                                   │   sec/op    │   sec/op     vs base                │
ForestSortedIteration/bbolt,root-8   207.2µ ± 6%   173.6µ ± 6%  -16.23% (p=0.000 n=10)
ForestSortedIteration/bbolt,leaf-8   3.910µ ± 5%   3.928µ ± 7%        ~ (p=0.529 n=10)
geomean                              28.46µ        26.11µ        -8.27%
```

They are not representative, as the worst case is when we have multiple
items of different lengths. However, `FileName` is usually less than 100
in practice, so the asymptotics is the same.

Signed-off-by: Evgenii Stratonikov <e.stratonikov@yadro.com>
This commit is contained in:
Evgenii Stratonikov 2024-08-23 10:58:33 +03:00
parent 6836d5df38
commit a38f26b481
3 changed files with 32 additions and 10 deletions

View file

@ -1161,6 +1161,7 @@ func (t *boltForest) fillSortedChildren(b *bbolt.Bucket, nodeIDs MultiNode, h *f
lastFilename = nil
nodes = nil
length = actualLength + 1
count = 0
c.Seek(append(prefix, byte(length), byte(length>>8)))
c.Prev() // c.Next() will be performed by for loop
}

View file

@ -237,9 +237,8 @@ func BenchmarkForestSortedIteration(b *testing.B) {
// The issue which we call "BugWithSkip" is easiest to understand when filenames are
// monotonically increasing numbers. We want the list of sorted filenames to have different length interleaved.
// The bug happens when we switch between length during listing.
// Thus this test contains numbers from 1 to 1000 and batch size of size 100.
// Thus this test contains numbers from 1 to 2000 and batch size of size 10.
func TestForest_TreeSortedIterationBugWithSkip(t *testing.T) {
t.Skip()
for i := range providers {
t.Run(providers[i].name, func(t *testing.T) {
testForestTreeSortedIterationBugWithSkip(t, providers[i].construct(t))

View file

@ -2,6 +2,8 @@ package pilorama
import (
"container/heap"
"slices"
"strings"
)
type heapInfo struct {
@ -29,6 +31,7 @@ func (h *filenameHeap) Pop() any {
// fixedHeap maintains a fixed number of smallest elements started at some point.
type fixedHeap struct {
start *string
sorted bool
count int
h *filenameHeap
}
@ -44,20 +47,39 @@ func newHeap(start *string, count int) *fixedHeap {
}
}
const amortizationMultiplier = 5
func (h *fixedHeap) push(id MultiNode, filename string) bool {
if h.start != nil && filename <= *h.start {
return false
}
heap.Push(h.h, heapInfo{id: id, filename: filename})
if h.h.Len() > h.count {
heap.Remove(h.h, h.h.Len()-1)
*h.h = append(*h.h, heapInfo{id: id, filename: filename})
h.sorted = false
if h.h.Len() > h.count*amortizationMultiplier {
slices.SortFunc(*h.h, func(a, b heapInfo) int {
return strings.Compare(a.filename, b.filename)
})
*h.h = (*h.h)[:h.count]
}
return true
}
func (h *fixedHeap) pop() (heapInfo, bool) {
if h.h.Len() != 0 {
return heap.Pop(h.h).(heapInfo), true
if !h.sorted {
slices.SortFunc(*h.h, func(a, b heapInfo) int {
return strings.Compare(a.filename, b.filename)
})
if len(*h.h) > h.count {
*h.h = (*h.h)[:h.count]
}
h.sorted = true
}
if len(*h.h) != 0 {
info := (*h.h)[0]
*h.h = (*h.h)[1:]
return info, true
}
return heapInfo{}, false
}