Performance Design

Performance is key to architecture design. The previous section describes scalability design. Scalability is a prerequisite for high performance. The following factors affect the performance of applications on the cloud:

• In terms of compute resources, latency, or the wait time between operations, stands out as the most straightforward metric for evaluating cloud computing efficiency.
• For network resources, throughput is the rate at which data is processed.
• Data transmission is represented by byte/second or bit/second. The throughput limit is an important performance bottleneck.
• For storage resources, IOPS is a measurement method of data transmission, which refers to the number of input/output operations per second.
• For database resources, the concurrency capability refers to the number of programs running in a period.

The following aspects also need to be considered: solution selection, performance measurement, performance monitoring, and performance balancing.

• Solution selection

  Select solutions tailored to different scenarios and combine multiple methods to meet requirements.

  Methods are continuously iterative and optimized and the data-driven method is also used to optimize the selection of resource types and configuration options.

• Performance measurement

  Set performance measurement and monitoring metrics to capture key performance metrics.

  Use visualization technologies to display performance metrics and issues (such as abnormal status and low utilization).

• Performance monitoring

  Determine the monitoring scope, measurement, and threshold.

  Create a complete view from multiple dimensions.

• Performance balancing

  Select a better solution considering the architecture to improve performance, such as using compression or caching.