This chapter presents metrics to quantify the degree of performance isolation a system provides. The metrics have been endorsed by the SPEC Research Group. In an ideal case, one should be able to measure performance isolation externally, that is, by running benchmarks from the outside and treating the system as a black box. This enables their use for a broad set of applications given that no internal knowledge of the system is required. The metrics and the thought process to create them serve as a practical example illustrating the metric attributes and principles introduced in Chapter 3. These metrics are applicable for use in performance benchmarks that measure performance without requiring internal knowledge. They are preferable in situations where different request sources use the functions of a shared system with a similar call probability and demand per request but with different load intensity. These characteristics are typical for multi-tenant applications but can also occur in other shared resource systems. The chapter introduces the metrics and provides a case study showing how they can be used in a real-life environment.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Performance Isolation

  • Rouven Krebs,
  • Samuel Kounev

摘要

This chapter presents metrics to quantify the degree of performance isolation a system provides. The metrics have been endorsed by the SPEC Research Group. In an ideal case, one should be able to measure performance isolation externally, that is, by running benchmarks from the outside and treating the system as a black box. This enables their use for a broad set of applications given that no internal knowledge of the system is required. The metrics and the thought process to create them serve as a practical example illustrating the metric attributes and principles introduced in Chapter 3. These metrics are applicable for use in performance benchmarks that measure performance without requiring internal knowledge. They are preferable in situations where different request sources use the functions of a shared system with a similar call probability and demand per request but with different load intensity. These characteristics are typical for multi-tenant applications but can also occur in other shared resource systems. The chapter introduces the metrics and provides a case study showing how they can be used in a real-life environment.