See what makes Loadzilla stand out from the crowd, like its powerful deployment model, in-depth error handling, analysis by consensus, seamless CI integration, historical trend analysis and more.Get started today with a free 14 day trial!
Get started
Loadzilla is a cloud platform that gives IT the added control and extra security of a dedicated server while giving users the convenience of SaaS. It also makes it massively scalable - the server can be deployed in any of 10 regions and depending on the chosen instance type it can handle up to 5,000 load agents supporting 2.5 million virtual users or more. Compare this to Blazemeter, which tops out at 5,000 virtual users.
What is called “continuous performance testing” invariably uses response times and error rates to define failure criteria. But error rates generate false positives, so it’s necessary to examine the test results manually. By contrast, Loadzilla prevents false positives by eliminating false negatives – a false negative is an error that isn’t really an error, one that is outside the team’s control, or one that can be safely ignored. Errors can be screened by type (400, 500, network, execution) and either excluded or ignored based on page, return code and response text. By eliminating false negatives, false positives are prevented, and the test will always pass unless an unexpected error or slowdown occurs.
The decision about what constitutes a false negative should be made by consensus, not a single individual. Users can be set up and assigned different roles, one of which is the ability to approve or reject test contracts. By adding users as approvers to a given test, they will be required to log in and approve or reject the test contract before it is used. At minimum the approvers should include someone in Dev and someone in QA. An audit log is maintained that includes each version of a contract, who approved it, and the associated test results.
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam.
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam.
Test results can be viewed in real time or after the fact in a Grafana Dashboard that shows all the relevant metrics in graph and tabular form. Top 10 Slowest Pages and Top 10 Variance are included in addition to the dashboard, and results can be viewed at both the summary and page level.
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam tempore qui quia dolor sed voluptatem inventore non facilis voluptatem aut
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam tempore qui quia dolor sed voluptatem inventore non facilis voluptatem aut quibusdam dolorem.
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam tempore qui quia dolor sed voluptatem inventore non facilis voluptatem aut quibusdam dolorem.
Lorem ipsum dolor sit amet. Sit numquam vitae vel minima ratione aut ipsum facilis. Qui ipsam tempore qui quia dolor sed voluptatem inventore non facilis voluptatem aut
The JMX and CSV files that comprise a test can be uploaded, or they can be pulled from GitHub. CSV files are maintained separately where they can be dynamically associated with any test. Version control for uploaded JMX and CSV files is maintained by the audit log feature. When a test is run with multiple agents the CSV file is automatically distributed between them
A JMeter test can be integrated into the build and deployment pipeline with a single line of code that executes a jar file on the build machine. The test can be run on the build machine, on one or more local machines, in the cloud, or any combination. After the test runs the results are uploaded and analyzed according to the test contract, returning a PASS or FAIL to the caller (e.g., Jenkins).
The test results page contains three performance gauges that show how each test performed compared to the previous test, a baseline, and an overall average. A tolerance value defines a range in which warnings are generated. These appear in yellow on the results page with failures appearing in red. Failure and warning notifications are sent to those who subscribe and can also be posted to a Slack or Teams channel.
For testing internally facing applications behind the firewall, the load agent can be installed on one or more virtual or physical machines. Up to 5,000 load agents can be deployed on AWS in regions worldwide.
Deploy the server of the size you need in the region closest to you. A dedicated server gives you extra control and added security while your users enjoy the convenience of SaaS.
Agree as a team about what response time and throughput limits and identify any false negatives that would otherwise cause the test to fail. This way it will always pass unless there’s an unexpected error or slowdown
Integrate with your build and deploy pipeline with a single line of code on the build machine, causing the test to be run and analyzed with the resulting PASS or FAIL status sent to the caller. Execute the test on the build machine, on one or more local machines, or any number of cloud instances.
Compare the results of each test to the previous one, a baseline and an average. Generate warnings if values get within a certain range. Quickly and easily spot trends over time.
Choose a plan that suits your business needs
FREE
Perfect plan to get started
A free plan grants you access to some cool features of Spend.In.
ULTIMATE
Best suits for great company!
If you a finance manager at big company, this plan is a perfect match.
PRO
Perfect plan for professionals!
For professional only! Start arranging your expenses with our best templates.
