Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hyj1991/v8-analytics
v8 engine's cpu & heap-memory analytics
https://github.com/hyj1991/v8-analytics
analytics bailout bailout-reason cpu cpu-profiler heap-memory heap-memory-analytics v8-profiler
Last synced: 8 days ago
JSON representation
v8 engine's cpu & heap-memory analytics
- Host: GitHub
- URL: https://github.com/hyj1991/v8-analytics
- Owner: hyj1991
- Created: 2017-01-16T05:17:26.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2017-07-01T14:46:23.000Z (over 7 years ago)
- Last Synced: 2024-05-22T01:21:39.107Z (6 months ago)
- Topics: analytics, bailout, bailout-reason, cpu, cpu-profiler, heap-memory, heap-memory-analytics, v8-profiler
- Language: JavaScript
- Homepage:
- Size: 567 KB
- Stars: 186
- Watchers: 5
- Forks: 15
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-nodejs - v8-analytics - V8 engine's cpu & heap-memory analytics.  (Repository / Performance Profiling/Analysis)
- awesome-github-star - v8-analytics - memory analytics | hyj1991 | 189 | (JavaScript)
README
[中文版](https://github.com/hyj1991/v8-cpu-analysis/blob/master/README_ZH.md)
[](https://badge.fury.io/js/v8-cpu-analysis)
[](http://packagequality.com/#?package=v8-cpu-analysis)
[](https://github.com/hyj1991/v8-cpu-analysis/LICENSE)
# v8-analytics/v8-cpu-analysis
Analysis the cpu & heap-memory log file which created by v8-profiler & heapdump，it can:
* **show you functions which are optimized failed by v8 engine**
* **show you functions which exectime greater than your expected**
* **show you suspicious memory leak point**
## Why Two Names？
```v8-analytics``` is completely equivalent to ```v8-cpu-analysis```, there is no difference between them.
Beacuse I want to introduce v8 engine's heap-memory analytics, so I think ```cpu``` in package name is not suitable. More, to be compatible with before, I chose them published comletely equivalent.
## Installation
### For Global
```bash
$ npm install v8-analytics -g
```
or
```bash
$ npm install v8-cpu-analysis -g
```
### For Embedded JS code
```bash
$ npm install v8-analytics
```
or
```bash
$ npm install v8-cpu-analysis
```
### Test
if you install -g, try:
```bash
//test bailout
$ va test bailout
$ va test bailout --only
//test timeout
$ va test timeout
$ va test timeout 200
$ va test timeout 200 --only
$ va test leak
```
```va test bailout --only``` can list you all functions which are deoptimized, and it's deoptimization reason.
```va test timeout 200 --only``` can list you all function which exectime > 200ms.
```va test leak``` cans list you suspicious leak point in your project.
## Quick Start
You can use this at the command line or embedded in your js code
### I. Command Line
#### Find Function Bailout Reason
The ```xxx.cpu.json``` file created by v8-profiler:
```bash
$ va bailout xxx.cpu.json
```
This will list all Function, and turn the deoptimized function to red. You can also use like that:
```bash
$ va bailout xxx.cpu.json --only
```
This will only list the deoptimized functions.
#### Find Funtion Exectime Greater Than Expected
```bash
$ va timeout xxx.cpu.json
```
This will list all Function, and their exectime
```bash
$ va timeout xxx.cpu.json 200
```
This will list all Function ,and turn the functions which exectime > 200ms to red.
```bash
$ va timeout xxx.cpu.json 200 --only
```
This will only list the functions which exectime > 200ms.
#### Find suspicious memory leak point
```
$ va leak xxx.mem.json
```
This will show you suspicous leak ponit in your node.js project.
### II. Embedded JS Code
```js
'use strict';
const fs = require('fs');
const v8Analytics = require('v8-analytics');
//or you can use following, they're equival
//const v8Analytics = require('v8-cpu-analysis');
//list all js function and it's execTime
const json = JSON.parse(fs.readFileSync('./test.cpu.json'));
const str = v8Analytics(json);
console.log(str);
//list you heap memory info
const json = JSON.parse(fs.readFileSync('./test.mem.json'));
const {leakPoint, heapMap, statistics} = analysisLib.memAnalytics(allData)
```