ibtree is an implementation of generic immutable balanced binary trees in Go. This packages provides generic immutable AVL trees.
Mutable btrees are all well and good, but sometimes you need a btree that can be updated and accessed at the same time in multiple different goroutines. Sure, you could invent a complicated locking scheme using ever more finegrained and deadlock prone locking schemes. There are libraries out there that provide that. This is not one of them.
Instead, this library provides immutable btrees. Any operation that would mutate the tree will instead make copies of any nodes that would be changed and return a new tree. The new tree and the old tree will share unchanged nodes. This library also provides bulk insert and delete operations that minimize the amount of node copying that happens under the hood.
go get https://github.com/VictorLowther/ibtree
package main
import github.com/VictorLowther/ibtree
import fmt
func main() {
tree := ibtree.New[int](func(a,b int) {return a < b})
tree = tree.Insert(0, 1, 2, 3, 4, 5, 6, 7, 8, 9)
tree = tree.Reverse()
iter := tree.Iterate(nil, nil)
for iter.Next() {
fmt.Println(iter.Item())
}
}
On a Macbook Pro M1 Max:
% go test -bench .
goos: darwin
goarch: arm64
pkg: github.com/VictorLowther/ibtree
BenchmarkInsertIntSeqNocow-10 6304920 214.8 ns/op 32 B/op 1 allocs/op
BenchmarkInsertIntSeqCow-10 7593422 176.1 ns/op 32 B/op 1 allocs/op
BenchmarkInsertIntSeqReverseNocow-10 6726402 225.3 ns/op 32 B/op 1 allocs/op
BenchmarkInsertIntSeqReverseCow-10 7352782 175.6 ns/op 32 B/op 1 allocs/op
BenchmarkInsertIntRandCow-10 2879599 731.1 ns/op 32 B/op 1 allocs/op
BenchmarkDeleteIntSeq-10 8109351 159.6 ns/op 32 B/op 1 allocs/op
BenchmarkDeleteIntRand-10 2684017 640.3 ns/op 32 B/op 1 allocs/op
BenchmarkInsertStringSeq-10 6578706 189.6 ns/op 48 B/op 1 allocs/op
BenchmarkInsertStringRand-10 1586451 933.3 ns/op 48 B/op 1 allocs/op
BenchmarkDeleteStringSeq-10 6984127 187.3 ns/op 48 B/op 1 allocs/op
BenchmarkDeleteStringRand-10 1548794 877.8 ns/op 48 B/op 1 allocs/op
BenchmarkIntIterAll-10 181415104 6.618 ns/op 0 B/op 0 allocs/op
BenchmarkFetch/btree_size_16-10 100000000 10.49 ns/op 0 B/op 0 allocs/op
BenchmarkFetch/map_size_16-10 201185232 6.035 ns/op
BenchmarkFetch/btree_size_256-10 50292850 21.89 ns/op 0 B/op 0 allocs/op
BenchmarkFetch/map_size_256-10 149541414 8.055 ns/op
BenchmarkFetch/btree_size_65536-10 11741773 105.2 ns/op 0 B/op 0 allocs/op
BenchmarkFetch/map_size_65536-10 61229953 19.14 ns/op
BenchmarkFetch/btree_size_16777216-10 1714615 691.6 ns/op 0 B/op 0 allocs/op
BenchmarkFetch/map_size_16777216-10 21412989 57.05 ns/op
PASS
ok github.com/VictorLowther/ibtree 86.026s
Interestingly enough, the slowdown on the random benchmarks appears to be due to branch misprediction rather than tree rebalancing performing more work -- dealing with sorted data actually performs more rebalancing than random data.
