Investigate how cascading failures occur across layers within your system architecture
Source: Pragmatic Bookshelf
The standard system architecture for enterprise systems, including Web sites and Web services, comprises a collection of functionally distinct farms or clusters (often referred to as layers) that are interconnected through some form of load balancing. When system failures start, they usually begin with a crack that comes from some fundamental problem. Various mechanisms can retard or stop the crack, but absent those mechanisms, the crack can progress and even be amplified by structural problems. In this sample chapter from Release It! Design and Deploy Production-Ready Software, investigate how a cascading failure occurs when a crack in one layer triggers a crack in a calling layer, and learn about the mechanisms that can be put in place to retard this failure.
Title: Release It! Design and Deploy Production-Ready Software
Published: March 2007
Authors: Michael T. Nygard
Chapter: Sample Anti-Patterns: Cascading Failures
Published by Pragmatic Bookshelf