Performance Overview

What is Performance in a System?

Performance in a system refers to how well the system operates under different conditions, including its speed, reliability, and ability to handle users or tasks. High-performing systems make sure users experience fast responses, smooth operations, and minimal errors, even when the workload is high. Performance is evaluated through measurable metrics that describe how efficiently a system works under normal or stressed conditions. These metrics help ensure the system meets user and business expectations.

Key Performance Metrics

Here’s a breakdown of the core performance elements:

1. Throughput (How much work the system can handle)

• What it means: Throughput measures the number of operations or tasks a system can complete within a specific time (e.g., number of requests it can process per second).

• Key Scenarios:

  • Normal Throughput: Performance under usual conditions.
  • Stressed Throughput: Performance under higher-than-average workloads.
  • Maximum Throughput: The highest workload a system can handle before slowing down or failing.

• Example:

  • “The system can handle 500 operations per second under normal conditions, 300 when stressed, and a maximum of 1000 operations.”

2. Latency (How fast the system responds)

• What it means: Latency is the time it takes for the system to respond to a request. Shorter latency means faster response.

• Key Metrics:

  • P95 Latency: The system responds faster than this time 95% of the time.
  • P99 Latency: The system responds faster than this time 99% of the time (more strict).
  • Degraded Latency: Response times under stress, when the system is working harder than usual.

• Example:

  • “Under normal conditions, 95% of responses are faster than 100ms. Even under stress, 95% of responses stay under 500ms.”

3. Error Rate (How often things go wrong)

• What it means: Error rate measures how frequently errors occur during system operations. Lower error rates mean the system is more reliable.

• Key Scenarios:

  • Normal Error Rate: Errors during usual conditions.
  • Stressed Error Rate: Errors during high-load conditions.

• Example:

  • “During normal operations, the error rate stays below 0.1%, but under stress, it can reach up to 2%.“

4. Resource Utilisation (How efficiently the system uses its resources)

• What it means: Resource utilisation measures how much of the system’s available hardware resources (like CPU and memory) are being used during operations.
• Why It Matters: High resource utilisation can affect system performance and lead to slowdowns or crashes.
• Example:
  • “CPU usage remains under 70% during normal operations, and memory usage stays under 75%.”

What Does Good Performance Look Like?

For a business or user, good performance means:

1. Fast Responses: The system responds instantly, or within acceptable time limits, to user actions or requests.
2. Handling Workloads: Whether one user or millions are accessing the system, it operates smoothly without lag or errors.
3. Minimal Downtime or Errors: Users encounter very few issues, even when the system is under heavy use.

Everyday Analogy

Imagine a highway:

• Throughput: How many cars can drive on the highway per minute.

  • During normal hours, the highway handles 500 cars per minute.
  • During rush hour, it handles 300 cars.
  • At maximum capacity, it can handle 1000 cars, but beyond that, traffic becomes congested.

• Latency: The time it takes for a car to get from one end of the motorway to the other.

  • During ideal traffic, drivers complete the trip in under 10 minutes (P95).
  • During rush hour, some drivers may take up to 30 minutes (degraded latency).

• Error Rate: The number of cars that break down or get into accidents on the motorway. Good road design, maintenance, and rules keep this number low.

• Resource Utilization: The condition of the motorway infrastructure (like lanes or traffic lights). Overuse of resources may result in cracks, delays, or breakdowns.

Why is Performance Important?

• For Users: A well-performing system means faster responses, fewer errors, and a smooth experience. For example, no one likes waiting for a website to load or a video to buffer.
• For Businesses: Good performance ensures customer satisfaction, prevents lost revenue, and maintains a competitive edge.

Summary

Performance ensures that systems:

1. Handle workloads properly (throughput).
2. Respond quickly (latency).
3. Minimize problems (error rate).
4. Use resources smartly (CPU/memory).

When these elements are optimised, users get a fast, reliable, and smooth experience, and businesses benefit from stable operations, even under heavy use!