Throughput is considered as one of the most confusing performance testing related term. People often struggle to understand this term and many of them end up ignoring it. In this article we are going to discuss about throughput in simplest language possible.

Jan 22, 2021 Apache JMeter is a 100% pure Java desktop application designed to load test functional behavior and measure performance. It was originally designed for testing Web Applications but has since expanded to other test functions. Apache JMeter The easiest way to get started with JMeter is to download the JMeter binaries. We need to be sure to grab the 3.2 version of the binaries: Another tip before launching JMeter is to verify the java version. This can be done in a command prompt using the java -version command.

Dec 13, 2012 Apache JMeter. Apache JMeter is open source software, a 100% pure Java desktop application designed to load test functional behavior and measure performance. JMeter is accessible to everybody, from the largest corporations down to individuals running their own consultancy.

So, what is throughput? As per the documentation throughput is requests/unit of time where the calculation of time begins from the start of the first sample to the end of the last one. Basically, it is the amount of transactions happened over test duration. It is also used to determine the capacity of the application under test. The ability of the server in terms of how much load it can take. Purpose of doing performance testing is to achieve highest level of throughput.

When presenting a performance test result, it is often expressed as transaction per second(TPS).

Throughput in Real Life Scenario

Let’s take an example of a Burger Joint. As illustrated below, the burger joint has three counters. Hence, at a time it can serve only 3 customers. Let’s assume that exact time to serve a customer is 1 minute, no matter how big or small the order is.

Therefore, no matter how many customers are there, only three customers this joint can entertain at a time. Remaining customers have to wait, thus lining up a queue. Throughput of Burger joint becomes 3(customer)/min. This is the upper limit i.e. maximum the joint can achieve.

Same concept is the case when we are testing a web application. If a web application receives 100 requests at the same time, but it is able to serve maximum 50 in one go, the remaining 50 ends up being in the queue. Number of concurrent threads and ramp up time have a direct impact on the throughput calculation. During the ramping up phase of performance test run, throughput may have lower amount compared to the phase where execution is in steady stage.

To analyze it in more detail we have to understand what is Latency in performance testing.

Latency

People always confuse themselves between Response Time and Latency. Suppose you are sending a request from the browser to a server. It takes certain amount of time to reach to the server. Application starts actual processing once the request reaches there. After the processing is over, response travels back to the browser.

In performance testing, the term latency of a request is the time request takes to travel from client to server and server to client. In other words, Delay. Hence,

Latency + Processing time(at application server level) = Response Time

The Response time, Latency and Throughput all are related with each other. Higher latency impacts on Response time and collectively they affects the Throughput value.

Throughput in Jmeter

Jmeter provides many ways to track the value of this useful parameter. Below are some Jmeter listeners you can use to analyze the Throughput value:

• Aggregate Graph
• Graph Results

Jmeter also provides a very powerful Constant Throughput Timer to set the constant value of TPS to test the application load. You can also use Throughput Controller to distribute the User load within single Thread Group.

Throughput is one of the significant indicator when determining the performance capacity of of the application server. On a high level it gives a rough idea about the number of transactions that can be made per given time frame. One should not entirely depend on the its value but consider it as one of the important factor.

• Idea Place
  

What Is JMeter? Introduction & Uses - Guru99

Apache JMeter

Apache JMeter is open source software, a 100% pure Java desktop application designed to load test functional behavior and measure performance.

JMeter is accessible to everybody, from the largest corporations down to individuals running their own consultancy. When used properly, it offers the ability to gain valuable insights into the performance of SAP BusinessObjects BI Platform, and can be used to maximize your return on investment (ROI) and lower your total cost of ownership (TCO).

Apache Jmeter Installation

Some discrete benefits of JMeter are:

• It may be used to test performance of both static resources such as JavaScript and HTML, as well as dynamic resources, such as JSP, Servlets, and AJAX.

• It can be used to simulate heavy load on an application to analyze overall performance under different types of load.

• It delivers a variety of graphical analyses of performance, such as graphs and charts, with a variety of results analysis tools available.

• It is Java based and open source, so provides a fully extensible platform for developers to continuously innovate on.

JMeter - Where and Why

Where to find JMeter

As part of the greater Apache project, JMeter benefits from a rich and knowledgeable community. You can download the tool from the JMeter Download Site and find a variety of useful tutorials and how-to articles on the official JMeter Wiki

Why use JMeter?

It is my firm belief that performance testing should be available to everybody. Without some form of benchmarking you cannot be confident that your BI Platform can handle the amount of usage you've estimated. Sizing calculators are a wonderful thing but they cannot take the specifics of your environment into consideration. You can read more about a general approach to performance testing in this SCN Blog

Now, there are many commercial performance testing tools on the market. Some of them, such as HP LoadRunner, are brilliant tools that make performance testing a fairly easy proposition for the people that can afford them. But what about the rest of us? As I mention above, performance testing should be available to everybody, and that's where JMeter really comes into play.

As free open source software, JMeter can be used by anybody. It does, however, require a fundamental understanding of how a particular application works. Web applications such as BI Platform create a unique user experience by using dynamic data that identifies a particular user, report, universe, or query. The large commercial performance testing tools are capable of automatically correlating this dynamic data for you and can simulate multiple users pretty easily.

JMeter doesn't have the same level of capability to automatically correlate dynamic data. JMeter was initially created for developers to load test their own applications, ones that they would naturally have an intimate familiarity with. Most people, unless they are SAP Developers, won't have the same level of familiarity with the BI Platform. The goal of this wiki is to provide you with the needed visibility into how BI Platform works 'under the hood', along with the building blocks for creating JMeter Test Plans, so that you can create rich, complex performance tests that make your BI Platform implementation run better.

Next Steps

Description of JMeter core components

Details regarding the components that comprise JMeter and make it work with SAP BusinessObjects BI Platform.

1st time setup of JMeter

Where to get it and how to configure it.

Creating your first Test Plan

The good stuff ... record a workflow and play it back to emulate business usage.

Additional considerations: Static vs. Dynamic Content

When to worry about static content like Javascript