Parallel Execution

Execute multiple HTTP requests simultaneously for improved performance.

Overview

Parallel execution allows curl-runner to send multiple HTTP requests simultaneously, significantly reducing the total execution time when testing multiple endpoints or performing load testing.

Note: By default, curl-runner executes requests sequentially. Enable parallel execution using the execution: parallel setting in your YAML file or the --parallel CLI flag.

Basic Usage

Configure parallel execution in your YAML file using the global execution setting.

parallel-requests.yaml

# Run multiple requests in parallel
global:
  execution: parallel
  variables:
    API_URL: https://api.example.com
  output:
    showMetrics: true

requests:
  - name: Get Users
    url: ${API_URL}/users
    method: GET
    
  - name: Get Posts
    url: ${API_URL}/posts
    method: GET
    
  - name: Get Comments
    url: ${API_URL}/comments
    method: GET
    
  - name: Get Products
    url: ${API_URL}/products
    method: GET

# Run multiple requests in parallel
global:
  execution: parallel
  variables:
    API_URL: https://api.example.com
  output:
    showMetrics: true

requests:
  - name: Get Users
    url: ${API_URL}/users
    method: GET
    
  - name: Get Posts
    url: ${API_URL}/posts
    method: GET
    
  - name: Get Comments
    url: ${API_URL}/comments
    method: GET
    
  - name: Get Products
    url: ${API_URL}/products
    method: GET

Configuration Options

Control how parallel execution behaves with these settings:

Setting	Type	Description
`execution`	string	"parallel" or "sequential" (default: "sequential")
`continueOnError`	boolean	Continue executing remaining requests if one fails
`timeout`	number	Global timeout for each request in milliseconds

Error Handling

When running requests in parallel, you can control how errors are handled using the continueOnError setting.

parallel-with-error-handling.yaml

# Control parallel execution
global:
  execution: parallel
  continueOnError: true  # Continue even if some requests fail
  variables:
    BASE_URL: https://api.example.com
  output:
    verbose: true
    showMetrics: true

requests:
  # These will all run at the same time
  - name: Request 1
    url: ${BASE_URL}/endpoint1
    timeout: 5000
    
  - name: Request 2
    url: ${BASE_URL}/endpoint2
    timeout: 5000
    
  - name: Request 3
    url: ${BASE_URL}/endpoint3
    timeout: 5000

# Control parallel execution
global:
  execution: parallel
  continueOnError: true  # Continue even if some requests fail
  variables:
    BASE_URL: https://api.example.com
  output:
    verbose: true
    showMetrics: true

requests:
  # These will all run at the same time
  - name: Request 1
    url: ${BASE_URL}/endpoint1
    timeout: 5000
    
  - name: Request 2
    url: ${BASE_URL}/endpoint2
    timeout: 5000
    
  - name: Request 3
    url: ${BASE_URL}/endpoint3
    timeout: 5000

Stop on Error

By default, if any request fails, curl-runner will cancel remaining requests

Continue on Error Recommended

Set continueOnError: true to complete all requests regardless of failures

Performance Testing

Parallel execution is ideal for performance testing and load simulation. Use it to test how your API handles concurrent requests.

performance-test.yaml

# Performance testing with parallel requests
global:
  execution: parallel
  variables:
    API_URL: https://api.example.com
    AUTH_TOKEN: your-token-here
  output:
    showMetrics: true
    saveToFile: performance-results.json

# Test multiple endpoints simultaneously
requests:
  - name: Health Check 1
    url: ${API_URL}/health
    
  - name: Health Check 2
    url: ${API_URL}/health
    
  - name: Health Check 3
    url: ${API_URL}/health
    
  - name: Health Check 4
    url: ${API_URL}/health
    
  - name: Health Check 5
    url: ${API_URL}/health

# Performance testing with parallel requests
global:
  execution: parallel
  variables:
    API_URL: https://api.example.com
    AUTH_TOKEN: your-token-here
  output:
    showMetrics: true
    saveToFile: performance-results.json

# Test multiple endpoints simultaneously
requests:
  - name: Health Check 1
    url: ${API_URL}/health
    
  - name: Health Check 2
    url: ${API_URL}/health
    
  - name: Health Check 3
    url: ${API_URL}/health
    
  - name: Health Check 4
    url: ${API_URL}/health
    
  - name: Health Check 5
    url: ${API_URL}/health

Metrics Collected

• Total execution time for all requests
• Individual request duration
• Response sizes
• Success/failure counts
• Average response time

CLI Usage

Control parallel execution from the command line.

terminal

# Run with parallel execution
curl-runner api-tests.yaml --parallel

# Override file setting to run sequentially
curl-runner api-tests.yaml --sequential

# Run parallel with verbose output
curl-runner api-tests.yaml --parallel --verbose

# Run with parallel execution
curl-runner api-tests.yaml --parallel

# Override file setting to run sequentially
curl-runner api-tests.yaml --sequential

# Run parallel with verbose output
curl-runner api-tests.yaml --parallel --verbose

CLI Flags

Flag	Description
`--parallel`	Force parallel execution
`--sequential`	Force sequential execution
`--continue-on-error`	Continue execution on failures

Best Practices

Use for Independent Requests

Parallel execution works best when requests don't depend on each other. If requests need data from previous responses, use sequential execution.

Set Appropriate Timeouts

When running many requests in parallel, set reasonable timeouts to prevent hanging requests from blocking completion.

Monitor Resource Usage

Running many parallel requests can consume significant system resources. Start with a small number and increase gradually.

Use Metrics for Analysis

Enable showMetrics: true to collect timing data for performance analysis.