CodeQuality • 2026-06-17

JMeter Performance Testing

Master JMeter performance testing with visual diagrams covering test setup, load testing, profiling, recording, processors, and scripting. Complete guide with workflows and best practices.

Q1: Performance Issue Resolution Process

flowchart TD
    Issue[Performance Issue Detected] --> Reproduce[Reproduce with JMeter]
    
    Reproduce --> Setup[Setup JMeter Test]
    Setup --> Concurrent[Configure Concurrent Users]
    Setup --> Load[Apply Heavy Load]
    
    Load --> Profile[Profile with jvisualvm]
    Profile --> Monitor[Monitor Metrics]
    
    Monitor --> CPU[CPU Times]
    Monitor --> Memory[Memory Usage]
    Monitor --> Threads[Thread Activity]
    Monitor --> GC[Garbage Collection]
    
    Monitor --> Identify[Identify Bottleneck]
    Identify --> Fix[Apply Fix]
    
    Fix --> Retest[Retest Performance]
    Retest --> Verify{Issue Resolved?}
    
    Verify -->|No| Profile
    Verify -->|Yes| Complete[Complete]
    
    style Issue fill:#FF5252
    style Complete fill:#4CAF50

Resolution Workflow:

Set up JMeter to reproduce production-like scenario with concurrent requests
Apply heavy load to system to expose performance bottlenecks
Use profiling tools (jvisualvm, JProfiler) to monitor CPU, memory, threads
Identify specific bottleneck (database, memory leak, inefficient code)
Common issues: connection leaks, backtracking regex, Cartesian SQL joins
Apply targeted fixes and retest to verify improvement

Q2: Performance Improvement Example

graph LR
    Before[Before<br/>Response Time: 4s] --> Analysis[Performance Analysis]
    
    Analysis --> Issues[Issues Found]
    Issues --> I1[DB Connection Leak]
    Issues --> I2[Backtracking Regex]
    Issues --> I3[Cartesian SQL Joins]
    
    Issues --> Fixes[Applied Fixes]
    Fixes --> F1[Fixed Connection Pool]
    Fixes --> F2[Optimized Regex]
    Fixes --> F3[Rewrote SQL Query]
    
    Fixes --> After[After<br/>Response Time: 2s]
    
    After --> Improvement[50% Improvement]
    
    style Before fill:#FF5252
    style After fill:#4CAF50
    style Improvement fill:#2196F3

Accomplishment Details:

Reduced response time from 4 seconds to 2 seconds (50% improvement)
Used JMeter to simulate peak load and identify performance degradation
Profiling revealed database connection leak causing resource exhaustion
Fixed backtracking regular expression causing CPU spikes
Rewrote SQL query to eliminate Cartesian joins reducing database load
Verified improvements under sustained load testing

Q3: JMeter Test Structure

graph TB
    TestPlan[Test Plan<br/>OrderExecution.jmx] --> ThreadGroup[Thread Group<br/>30 Concurrent Users]
    
    ThreadGroup --> Config[Configuration Elements]
    Config --> Variables[User Defined Variables<br/>HOST, PORT, DATA_PATH]
    Config --> CSV[CSV Data Config<br/>TestData.csv]
    Config --> Headers[HTTP Header Manager<br/>Content-Type: application xml]
    Config --> Defaults[HTTP Request Defaults<br/>Server, Protocol]
    
    ThreadGroup --> Controllers[Controllers]
    Controllers --> Loop[Loop Controller<br/>Iterations: 1]
    
    Loop --> Samplers[Samplers]
    Samplers --> HTTP[HTTP Request<br/>POST Order XML]
    
    HTTP --> Assertions[Assertions]
    Assertions --> XPath[XPath Assertion<br/>Verify Response]
    
    HTTP --> Timers[Timers]
    Timers --> Gaussian[Gaussian Timer<br/>3 min delay]
    
    HTTP --> Listeners[Listeners]
    Listeners --> Results[View Results Tree]
    Listeners --> Graph[Graph Results]
    Listeners --> Summary[Summary Report]
    
    style TestPlan fill:#FF9900
    style ThreadGroup fill:#4CAF50

Test Components:

Thread Group: Simulates 30 concurrent users with 180s ramp-up (3 minutes)
User Defined Variables: Centralize configuration (HOST, PORT, paths) for easy modification
CSV Data Config: Read test data from file, cycle through different account IDs
HTTP Request Defaults: Set common parameters to avoid repetition in each request
Samplers: Execute actual HTTP requests with dynamic file paths using variables
Listeners: Collect and display test results (tree view, graphs, summary reports)

Q4: JMeter Recording Methods

sequenceDiagram
    participant Browser
    participant Proxy as JMeter Proxy<br/>Port 9090
    participant JMeter as JMeter WorkBench
    participant TestPlan as Test Plan
    
    Note over Browser,TestPlan: Method 1: JMeter as Proxy
    
    Browser->>Proxy: Configure Browser<br/>Proxy: localhost:9090
    Browser->>Proxy: Browse Application<br/>http://myhost:7001/myapp
    Proxy->>JMeter: Record HTTP Requests
    JMeter->>JMeter: Filter Requests<br/>Include: .*myapp.*<br/>Exclude: .*\.jpg
    JMeter->>TestPlan: Move Recorded Elements
    
    Note over Browser,TestPlan: Method 2: BadBoy Tool
    
    Browser->>Browser: Use BadBoy to Record
    Browser->>Browser: Export as .jmx file
    Browser->>TestPlan: Import into JMeter
    TestPlan->>TestPlan: Clean up and extend

Recording Approaches:

JMeter Proxy: Configure JMeter as HTTP proxy server on port 9090
Set browser to use JMeter proxy, then browse application to record actions
Use include/exclude patterns to filter relevant requests (avoid images, CSS)
BadBoy Tool: External tool for recording web interactions, exports to .jmx format
Generated scripts require cleanup but provide good starting point
Add user-defined variables, CSV config, and assertions after recording

Q5: JMeter Element Execution Order

flowchart TD
    Start[Test Execution Starts] --> Config[0. Configuration Elements<br/>Variables, Defaults, CSV]
    
    Config --> PreProc[1. Pre-Processors<br/>Modify Request Before Execution]
    
    PreProc --> Timer[2. Timers<br/>Delay Before Request]
    
    Timer --> Sampler[3. Sampler<br/>Execute HTTP Request]
    
    Sampler --> Check{SampleResult<br/>Not Null?}
    
    Check -->|Yes| PostProc[4. Post-Processors<br/>Extract Values from Response]
    Check -->|No| End[Skip Remaining]
    
    PostProc --> Assert[5. Assertions<br/>Validate Response]
    
    Assert --> Listen[6. Listeners<br/>Record Results]
    
    Listen --> Next{More<br/>Iterations?}
    
    Next -->|Yes| Config
    Next -->|No| Complete[Test Complete]
    
    style Config fill:#E3F2FD
    style Sampler fill:#FF9900
    style Complete fill:#4CAF50

Execution Sequence:

Configuration Elements: Set up variables, defaults, and data sources first
Pre-Processors: Modify sampler settings before request execution
Timers: Introduce delays to simulate realistic user behavior
Samplers: Execute actual requests (HTTP, JDBC, JMS, etc.)
Post-Processors: Extract values from response (regex, XPath) for subsequent requests
Assertions: Validate response content, status codes, response times

Q6: Pre-Processor and Post-Processor Usage

graph TB
    subgraph PreProcessor[Pre-Processor Use Cases]
        Pre[Pre-Processor] --> P1[Modify Request Parameters]
        Pre --> P2[Update Variables]
        Pre --> P3[Set HTTP Headers]
        Pre --> P4[Prepare Request Body]
    end
    
    Sampler[HTTP Sampler<br/>Execute Request] --> Response[Response Received]
    
    subgraph PostProcessor[Post-Processor Use Cases]
        Post[Post-Processor] --> PP1[Extract Values<br/>Regex Extractor]
        Post --> PP2[Parse Response<br/>XPath Extractor]
        Post --> PP3[Extract Tokens<br/>jsfViewState, CID]
        Post --> PP4[Store for Next Request]
    end
    
    PreProcessor --> Sampler
    Response --> PostProcessor
    
    style PreProcessor fill:#E3F2FD
    style PostProcessor fill:#FFF3E0
    style Sampler fill:#FF9900

Processor Functions:

Pre-Processor: Executes before sampler, modifies request settings or updates variables
Common use: Set dynamic headers, modify request parameters based on previous responses
Post-Processor: Executes after sampler, processes response data
Regular Expression Extractor: Extract values using regex patterns for correlation
XPath Extractor: Parse XML/HTML responses to extract specific elements
Example: Extract jsfViewState and CID from JSF/Seam framework responses for subsequent requests

Q7: BeanShell Scripting for Flexibility

graph TB
    JMeter[JMeter Test] --> BeanShell[BeanShell Scripting]
    
    BeanShell --> Features[Features]
    Features --> F1[Embedded Java Interpreter]
    Features --> F2[Custom Functions]
    Features --> F3[File Operations]
    Features --> F4[Complex Logic]
    
    BeanShell --> Examples[Use Cases]
    Examples --> E1[Extract and Write Order IDs]
    Examples --> E2[Date Calculations<br/>getReportStartDate]
    Examples --> E3[Conditional Logic<br/>isTenth loopCounter]
    Examples --> E4[Custom Validations]
    
    BeanShell --> Integration[Integration]
    Integration --> I1[BeanShell PostProcessor]
    Integration --> I2[BeanShell PreProcessor]
    Integration --> I3[Custom Functions<br/>MyBeanShell.bsh]
    Integration --> I4[Built-in Functions<br/>__time, __Random]
    
    style BeanShell fill:#FF9900
    style Features fill:#4CAF50

BeanShell Capabilities:

Embedded Java interpreter for custom scripting within JMeter
Write complex logic not possible with standard JMeter elements
Extract values from responses and write to files for analysis
Create reusable custom functions in .bsh files stored in JMETER_HOME/bin
Invoke functions using ${__BeanShell(return myFunction();)} syntax
Alternative: Write custom Java functions implementing org.apache.jmeter.functions.Function interface

Q8: JMeter Reusability Best Practices

graph TB
    Reusability[JMeter Reusability] --> Config[Configuration Reuse]
    Reusability --> Modular[Modularization]
    Reusability --> Scripting[Custom Scripting]
    
    Config --> C1[CSV Data Set Config<br/>External Test Data]
    Config --> C2[User Defined Variables<br/>Centralized Settings]
    Config --> C3[HTTP Request Defaults<br/>Common Parameters]
    Config --> C4[Property Files<br/>Environment Config]
    
    Modular --> M1[Module Controller<br/>Reusable Test Fragments]
    Modular --> M2[Test Fragment<br/>Shared Login Logic]
    Modular --> M3[Include Controller<br/>External Test Plans]
    
    Scripting --> S1[BeanShell Functions<br/>Custom Utilities]
    Scripting --> S2[JSR223 Sampler<br/>Groovy Scripts]
    Scripting --> S3[Function Helper<br/>Reusable Functions]
    
    style Reusability fill:#FF9900
    style Config fill:#4CAF50
    style Modular fill:#2196F3

Reusability Strategies:

CSV Data Set Config: Externalize test data for easy modification without changing scripts
User Defined Variables: Centralize configuration (URLs, credentials) for environment switching
Module Controller: Create reusable test fragments (login, logout) invoked from multiple tests
BeanShell Functions: Write custom functions once, reuse across multiple test plans
Test Fragments: Define shared logic (authentication, common workflows) referenced by multiple thread groups
Avoid hardcoding values—use variables and properties for flexibility