Add benchmark test and results

README.md

@@ -88,3 +88,7 @@ $ go test -v ./...
 ## Credits
 
 - Uses 64 bit [xxHash](https://cyan4973.github.io/xxHash/) algorithm, an extremely fast non-cryptographic hash algorithm, for creating the hash. Implementations from [cespare](https://github.com/cespare/xxhash)
+
+## [Benchmarks](/benchmark)
+
+- It can compare 10 Million CSV records in under 2 minutes.

benchmark/README.md

@@ -0,0 +1,19 @@
+## Benchmark Results
+
+Benchmark test can be found [here](https://github.com/aswinkarthik93/csvdiff/blob/master/pkg/digest/digest_benchmark_test.go).
+
+```bash
+$ cd ./pkg/digest
+$ go test -bench=. -v -benchmem -benchtime=5s -cover
+```
+
+|                              |            |                         |                      |                     |
+| ---------------------------- | ---------- | ----------------------- | -------------------- | ------------------- |
+| BenchmarkCreate1-8           |    2000000 |             5967 ns/op  |           5474 B/op  |        21 allocs/op |
+| BenchmarkCreate10-8          |     500000 |            16251 ns/op  |          10889 B/op  |        94 allocs/op |
+| BenchmarkCreate100-8         |     100000 |           114219 ns/op  |          67139 B/op  |       829 allocs/op |
+| BenchmarkCreate1000-8        |      10000 |          1042723 ns/op  |         674239 B/op  |      8078 allocs/op |
+| BenchmarkCreate10000-8       |       1000 |         10386850 ns/op  |        6533806 B/op  |     80306 allocs/op |
+| BenchmarkCreate100000-8      |        100 |        108740944 ns/op  |       64206718 B/op  |    804208 allocs/op |
+| BenchmarkCreate1000000-8     |          5 |       1161730558 ns/op  |       672048142 B/op |  8039026 allocs/op  |
+| BenchmarkCreate10000000-8    |          1 |       12721982424 ns/op |       6549111872 B/op| 80308455 allocs/op  |

pkg/digest/digest_benchmark_test.go

@@ -0,0 +1,61 @@
+package digest
+
+import (
+	"fmt"
+	"io"
+	"testing"
+)
+
+const SomeText = "something-name-%d,346345ty,fdhfdh,5436456,gfgjfgj,45234545,nfhgjfgj,45745745,djhgfjfgj"
+
+func BenchmarkCreate1(b *testing.B)     { benchmarkCreate(1, b) }
+func BenchmarkCreate10(b *testing.B)    { benchmarkCreate(10, b) }
+func BenchmarkCreate100(b *testing.B)   { benchmarkCreate(100, b) }
+func BenchmarkCreate1000(b *testing.B)  { benchmarkCreate(1000, b) }
+func BenchmarkCreate10000(b *testing.B) { benchmarkCreate(10000, b) }
+
+func BenchmarkCreate100000(b *testing.B)   { benchmarkCreate(100000, b) }
+func BenchmarkCreate1000000(b *testing.B)  { benchmarkCreate(1000000, b) }
+func BenchmarkCreate10000000(b *testing.B) { benchmarkCreate(10000000, b) }
+
+func benchmarkCreate(limit int, b *testing.B) {
+	for i := 0; i < b.N; i++ {
+		CreateDigestFor(limit, b)
+	}
+}
+
+func CreateDigestFor(count int, b *testing.B) {
+	b.StopTimer()
+	reader := &Reader{limit: count}
+
+	config := &Config{
+		Reader: reader,
+		Key:    []int{0},
+		Value:  []int{1},
+	}
+
+	b.StartTimer()
+	Create(config)
+}
+
+type Csv struct {
+	counter int
+	limit   int
+}
+
+type Reader struct {
+	counter int
+	limit   int
+}
+
+func (r *Reader) Read(p []byte) (n int, err error) {
+	if r.counter == r.limit {
+		return 0, io.EOF
+	}
+	toRead := fmt.Sprintf("%d,%s\n", r.counter, SomeText)
+	r.counter++
+	for i, b := range []byte(toRead) {
+		p[i] = b
+	}
+	return len(toRead), nil
+}