spring-banq

Banq is a lightweight runtime performance measurement tool based on Spring AOP (Aspect-Oriented Programming). It allows you to measure execution times of methods in Spring-managed beans simply by annotating them with @Banq.

Note

⚠️ Important: Banq only works on methods called through Spring’s proxy mechanism. That means the method must belong to a Spring-managed bean. Direct self-calls inside the same class will not be intercepted.

Build

Java 19+ required. mvn clean install package

Setup

1. Add to Project

Add the two classes to your project source folder: /src/main/de/teatime08

2. Add the Annotation

Mark any method in a Spring @Component, @Service, or @Controller class with the @Banq annotation:

import de.teatime08.Banq;
import org.springframework.stereotype.Service;

@Service
public class ExampleService {

    @Banq
    public String doWork() {
        // some business logic
        return "done!";
    }
}

3. Spring Context Scan

The performance measuring logic is already provided by the BanquetteAspect class. Simply make sure it is part of your Spring application context:

import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;

@SpringBootApplication(scanBasePackages = "de.teatime08")
public class Application {
    public static void main(String[] args) {
        SpringApplication.run(Application.class, args);
    }
}

The @Aspect and @Component annotations on BanquetteAspect will ensure it is picked up automatically.

Measure Results

1. Collect Measurements

Whenever a @Banq method is executed, Banq records:

• Count: The running number of the measurement.
• Tick Date: The timestamp when the execution finished.
• Execution Time: Duration in nanoseconds.

All measures are stored in memory inside a Map<String, LinkedList>, keyed by the method signature. You can access them programmatically:

@Autowired
private BanquetteAspect banquetteAspect;

public void printMeasures() {
    System.out.println(banquetteAspect.printToString());
}

2. Statistical Outputs

Banq isn’t just logging times. It provides simple statistical helpers:

• Median:

double median = banquetteAspect.median("String de.teatime08.ExampleService.doWork()");

• Percentiles (e.g., 90th):

double p90 = banquetteAspect.percentile("String de.teatime08.ExampleService.doWork()", 90);

BoxPlot summary (min, max, median, Q1, Q3):

BanquetteAspect.BoxPlot stats = banquetteAspect.boxPlot("String de.teatime08.ExampleService.doWork()");
System.out.println(stats);

3. Example Output

If you’ve run your @Banq-annotated method several times, calling printToString() might give you something like:

String de.teatime08.ExampleService.doWork() BanqTick{count=0, tickDate=2025-10-03T12:01:05, benchNanoTime=105321}
String de.teatime08.ExampleService.doWork() BanqTick{count=1, tickDate=2025-10-03T12:01:07, benchNanoTime=100842}
String de.teatime08.ExampleService.doWork() BanqTick{count=2, tickDate=2025-10-03T12:01:09, benchNanoTime=99054}

This makes it easy to plug into logging, monitoring dashboards, or just use for ad-hoc performance checks during development.

Note 2

• Works only with methods in Spring-managed beans (@Component, @Service, @Controller, etc.).
• Results are stored in memory — no persistence layer is included.
• The statistics are simple by design (median, percentiles, boxplot). If you need advanced metrics or visualization, export them to a monitoring system.

Overall this is a small tool to annotate, run, measure, analyze the performance of smaller codeblocks in your Spring application.