Generic faultloads based on software faults for dependability benchmarking

摘要: The most critical component of a dependability benchmark is the faultload, as it should represent repeatable, portable, representative, and generally accepted set faults. These properties are essential to achieve desired standardization level required by but, unfortunately, very hard achieve. This particularly true for software faults, which surely accounts fact that this important class faults has never been used in known proposals. paper proposes new methodology definition faultloads based on benchmarking. Faultload such repeatability, portability scalability also analyzed validated through experimentation using case study benchmarking Web-servers. We concluded fault-based generated our appropriate useful As not tied any specific vendor or platform, can be generate evaluation product OLTP systems.