benchmarker

package module
v0.0.11 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Nov 4, 2024 License: Apache-2.0 Imports: 17 Imported by: 0

README

benchmarker

Simple benchmarker written in go.

Demo

func TestStartBenchmark(t *testing.T) {
	_, _, _ = benchmarker.StartBenchmark(benchmarker.BenchmarkSpec{
		Concurrent: 3,
		Round:      10,
		BuildReqFunc: func() (*http.Request, error) {
			return http.NewRequest(http.MethodGet, "http://localhost:8080", nil)
		},
	})
}

If you need CLI support:

//   -conc int
//         Concurrency (default 1)
//   -concgroup string
//         Concurrency Groups (e.g., '1,30,50', is equivalent to running the benchmark three times with concurrency 1, 30 and 50)
//   -debug
//         Enable debug log
//   -dur duration
//         Duration
//   -round int
//         Round (default 2)
func main() {
	_, err := benchmarker.StartBenchmarkCli(benchmarker.BenchmarkSpec{
		BuildReqFunc: func() (*http.Request, error) {
			return http.NewRequest(http.MethodGet, "http://localhost:8080", nil)
		},
	})
	if err != nil {
		panic(err)
	}
}

// concurrency 3, duration 10 seconds
// go run main.go -dur 10s -conc 3

Output

Benchmark Time: 2024/11/04 10:02:40 (CST)

--------- Brief ---------------

total_time: 19.330292ms
total_requests: 30
throughput: 1552 req/sec
concurrency: 3
rounds (for each worker): 10
status_count: map[200:30]
success_count: map[true:30]

--------- Latency -------------

min: 297.208µs
max: 7.5815ms
median: 1.639542ms
avg: 1.787877ms
P75: 2.096292ms
P90: 2.976333ms
P95: 3.4655ms
P99: 7.5815ms

--------- Data ----------------

data file: benchmark_records.txt

-------------------------------

Timestamp: 1730685760316608, Took: 7.5815ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760316629, Took: 2.976333ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760316649, Took: 850.584µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760317501, Took: 2.607125ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760319606, Took: 1.073083ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760320109, Took: 1.7545ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760320680, Took: 3.4045ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760321864, Took: 2.914584ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760324085, Took: 1.837791ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760324190, Took: 3.4655ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760324779, Took: 454.208µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760325233, Took: 1.886625ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760325924, Took: 2.74725ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760327120, Took: 2.096292ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760327655, Took: 1.800208ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760328672, Took: 1.266667ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760329217, Took: 1.027958ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760329456, Took: 1.121542ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760329938, Took: 940.25µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760330245, Took: 1.696792ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760330578, Took: 1.404125ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760330879, Took: 1.582292ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760331942, Took: 638.625µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760331982, Took: 1.978917ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760332462, Took: 1.826042ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760333961, Took: 1.069208ms, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760334288, Took: 736.417µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760335031, Took: 297.208µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760335328, Took: 301.833µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]
Timestamp: 1730685760335630, Took: 298.375µs, Success: true (100.00%), HttpStatus: 200, Extra: map[]

Plots

plot_sorted_by_latency.png

plot_sorted_by_request_order.png

plot_success_rate.png

Documentation

Index

Constants

View Source 
const (
	// rough estimate on how many benchmark results will be created by one goroutine, increase it if necessary.
	DefaultResultQueueSize = 1000
)

Variables

This section is empty.

Functions

func StartBenchmark 

func StartBenchmark(spec BenchmarkSpec) ([]Benchmark, Stats, error)

Types

type Benchmark 

type Benchmark struct {
	Timestamp  int64
	Took       time.Duration
	Success    bool
	Extra      map[string]any
	HttpStatus int
	// contains filtered or unexported fields
}

func SortTimestamp 

func SortTimestamp(bench []Benchmark) []Benchmark

func SortTook 

func SortTook(bench []Benchmark) []Benchmark

type BenchmarkSpec added in v0.0.2 

type BenchmarkSpec struct {
	Concurrent int
	Round      int
	Duration   time.Duration

	// required, func to build benchmark request
	BuildReqFunc BuildRequestFunc

	// optional, by default, it considers 200 as a success.
	ParseResFunc ParseResponseFunc

	// funcs to log extra statistics information
	LogStatFunc []LogExtraStatFunc

	DebugLog          bool
	DisablePlotGraphs bool

	// do not include min/max labels on graph
	DisablePlotInclMinMaxLabels bool

	// do not draw percentile lines on graph
	DisablePlotInclPercentileLines bool

	// do not write benchmark records to file
	DisableOutputFile bool

	// rough estimate on how many benchmark results will be created by one worker, by default 1000.
	SingleWorkerResultQueueSize int

	PlotWidth                        font.Length
	PlotHeight                       font.Length
	PlotSortedByRequestOrderFilename string
	PlotSortedByLatencyFilename      string
	PlotSuccessRateFilename          string
	DataOutputFilename               string
	// contains filtered or unexported fields
}

type BuildRequestFunc 

type BuildRequestFunc func() (*http.Request, error)

type CliBenchmarkResult added in v0.0.6 

type CliBenchmarkResult struct {
	Benchmarks []Benchmark
	Stats      Stats
}

func StartBenchmarkCli added in v0.0.5 

func StartBenchmarkCli(spec BenchmarkSpec) ([]CliBenchmarkResult, error)

type LogExtraStatFunc 

type LogExtraStatFunc func([]Benchmark) string

type ParseResponseFunc 

type ParseResponseFunc func(buf []byte, statusCode int) Result

type Percentile added in v0.0.3 

type Percentile struct {
	Record Benchmark
	Index  int // index of record when they are sorted by time duration.
}

type Result 

type Result struct {
	HttpStatus int
	Success    bool
	Extra      map[string]any
}

type SendRequestFunc 

type SendRequestFunc func(c *http.Client) Result

type Stats 

type Stats struct {
	TotalTime     time.Duration
	TotalRequests int
	Throughput    float64
	StatusCount   map[int]int
	SuccessCount  map[bool]int
	Min           time.Duration
	Max           time.Duration
	Avg           time.Duration
	Med           time.Duration
	Percentiles   map[int]Percentile
}

func (*Stats) PercentileString added in v0.0.10 

func (s *Stats) PercentileString() string

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.