Alibaba Cloud Performance Testing (PTS) is a commercial performance testing tool that allows you to initiate stress testing tasks on demand. PTS scales to millions of concurrent virtual users (VUs) and tens of millions of transactions per second (TPS), and is fully compatible with JMeter.
Developers, QA engineers, and SREs use PTS to:
Validate that applications meet performance targets before launch
Identify bottlenecks and system limits under controlled, realistic load
Plan capacity by measuring throughput, response time, and error rates at scale
Perform end-to-end stress testing across business systems
What you can do with PTS
| Capability | Description |
|---|---|
| Scenario orchestration | Build multi-step test scenarios with API chaining, conditional logic, and parameterized data |
| API debugging | Test and validate individual API requests before running a full load test |
| Traffic customization | Shape load profiles by concurrency ramp-up, geographic distribution, and request rate |
| Real-time monitoring | Track response time, throughput, error rate, and resource utilization while a test runs |
| Stress testing reports | Generate post-test reports with performance metrics, trends, and bottleneck analysis |
How PTS works
Configure your test. In the PTS console, define the APIs to test, build a stress testing scenario, and set the load mode and target concurrency. Start, stop, or adjust load at any time.
Run the test. The stress testing control center automatically schedules load generators and distributes test data. Load generators randomly schedule Alibaba Cloud CDN points of presence (POPs) across hundreds of cities and carrier networks worldwide to initiate stress testing on specified business websites. The resulting traffic is geographically dispersed and closely resembles real user behavior.
Monitor results in real time. Live metrics appear in the PTS console while the test runs. If integrated, CloudMonitor and Application Real-Time Monitoring Service (ARMS) together with PTS collect stress testing data in real time based on various monitoring metrics.
Analyze the report. After the test completes, PTS generates a stress testing report. Use the performance data to identify performance issues and discover system bottlenecks based on the performance data of the stress testing scenario.
Test creation methods
PTS supports two ways to create stress testing scenarios. Start with visualized orchestration for a no-code approach, or upload a JMeter script for full scripting flexibility.
Visualized orchestration (recommended)
Build test scenarios through a visual interface in the PTS console, without scripting or coding. PTS uses its own high-performance load generators to run the test.
JMeter script upload
Upload an existing Apache JMeter test plan (.jmx) directly to PTS. PTS runs the script on a native JMeter load generator and adds custom load construction and monitoring data aggregation. For details, see Create a JMeter scenario.
PTS load generator advantages
The PTS in-house load generator provides capabilities beyond standard JMeter execution:
| Capability | Description |
|---|---|
| API-level RPS throughput modes | Control load by requests per second at the individual API level |
| Second-level startup | Ramp load generators to full capacity within seconds |
| Real-time control | Adjust test parameters while a test runs |
| Scheduled stress testing | Trigger tests automatically at predefined times |
| Global traffic distribution | Spread test traffic across carrier networks worldwide for geographically realistic load patterns |
Get started
| Path | Description |
|---|---|
| PTS V3.0 usage guide | Step-by-step learning path from basics to advanced topics |
| Create a JMeter scenario | Upload and run an existing JMeter test plan in PTS |