Скачать или смотреть Performance Testing Terminologies

Throughput: The amount of data transferred over a network or processed by an application in a given period, often measured in requests per second or bits per second.

Response Time: The time taken by an application to respond to a user request, measured from the time a request is sent until the first response byte is received.

Latency: The delay between a request and the start of the response, often referred to as "network delay" or "time to first byte (TTFB)."

Load Testing: Testing to determine how a system behaves under a specific expected load, typically to identify bottlenecks or slowdowns.

Stress Testing: Testing to evaluate how a system performs under extreme load conditions, often exceeding its normal capacity to assess its breaking point.

Spike Testing: A type of stress test that assesses system performance when subjected to sudden, extreme increases in load over a short period.

Endurance (Soak) Testing: Testing to ensure a system can handle a continuous expected load over an extended period to reveal issues like memory leaks or degradation over time.

Scalability Testing: Testing to determine the application's ability to scale up or down in response to increased load.

Bottleneck: A point in the system where performance slows down due to a particular resource constraint (e.g., CPU, memory, network bandwidth).

Concurrent Users: The number of users actively interacting with the system simultaneously during testing.

TPS (Transactions Per Second): The number of transactions completed per second, a common metric for measuring system performance.

RPS (Requests Per Second): The number of HTTP or API requests a server can handle per second.

Error Rate: The percentage of requests that fail due to errors, often monitored to assess application stability under load.

Think Time: The pause between user actions, typically simulated to mimic real-world user behavior during performance tests.

Warm-Up Period: The initial phase in load testing where load is gradually increased to prevent sudden spikes, allowing systems to reach steady-state performance.

Baseline Testing: Measuring the performance of a system under normal conditions as a benchmark for future comparison.

Workload Model: A model defining how users interact with an application, including the distribution and type of requests made during testing.

Load Generator: The tool or server used to simulate user load on an application during performance testing.

Think Time: Simulated pauses between requests to emulate real user behavior, used in test scripts to generate realistic load patterns.

Capacity Testing: Testing to determine the maximum number of users or transactions the system can handle before performance degrades.