Akamai CloudTest Reviews

I use the solution in my company for load testing. You can say that it is used on the API and then for web page-level load testing.

The most valuable features of the solution are that it is user-friendly and offers live monitoring, which gives you more data. On the fly, you can create custom dashboards and have response time breakup between your core HTML and non-HTML areas, like CSS.

We use this solution mostly for generating loads for a couple of our retail clients, especially for holiday readiness. We want to get our system up and running to get ready for unpredicted loads. So, we use it more for load testing.

In terms of improvement, I think integration of these tools with the leading EPM tools would be good. It would help to seamlessly integrate to Dynatrace or AppDynamics to understand what the profiling looks like when generating a load.

CI/CD integration, the pipeline itself, still needs to be improved.

Resource consumption also needs to be worked on. When we generate a thousand or two thousand concurrent users, we need to optimize the load generator resource utilization. This is because you need multiple servers in order to generate heavy loads. Other solutions like Gatling take much lower resources compared to that by Akamai CloudTest. For instance, you would need three different instances of typical 8-core 16 GB to generate 2000 concurrent users from Akamai CloudTest, but you would need half of the capacity with Gatling. 

Our primary use case is to do performance testing and load testing to determine which applications we release on the market to end-users. Before they go to the production environment, we perform a couple of tests to determine the threshold or maximum capacity our backend can handle. So we look at the capacity of our infrastructure and see if our backend can expand, how much it expands, where it goes, how it scales and if it impacts the response time. We can test all these features using Akamai CloudTest.

Since we run a lot of marketing campaigns on our products, we don't have the exact user capacity requirements. After running the load test we can determine the capacity our backend can handle. Based on that, we can make sure that our infrastructure is optimized. It helps us to cater to on-demand capacity and load, which helps to keep in mind the response time or the performance of applications for the end-users. We don't move it during peak hours, because our response time in applications become slower and unresponsive, or they simply behave differently. 

The agents on the cloud have been extremely valuable for us.

The test clip should be more user-friendly.

Soasta CloudTest is an integral part of our performance testing strategy and key in generating load across geography.

This is an awesome performance testing tool for web based applications, able to generate load multiple geographies, dynamic ramp-up to any levels of virtual users.

Before moving to Soasta CloudTest, we had a performance testing tool deployed on-premise environment. Other than high licensing cost, this tool has limitation in simulating real user experience to generate load in different geographies, hence, reliable performance test results. 

Soasta CloudTest has added flexility in performance testing plan, achieving at low cost.

I have only used CloudTest and I consider it a very powerful tool.

I haven’t used it outside the CloudTest training. It is an option for Performance testing taken into consideration by my company.

• Cloud load agents on-demand integration with AWS

The cloud-based load test is a realistic load test.