Failure Trace Archive

Welcome to the webpage of the Failure Trace Archive.

The Failure Trace Archive (FTA) is centralized public repository of availability traces of distributed systems, and tools for their analysis. The purpose of this archive is to facilitate the design, validation, and comparison of fault-tolerant models and algorithms.

In particular, the FTA contains the following:

• availability traces of distributed systems, differing in scale, volatility, and usage
• scripts and tools for analyzing these traces
• links to complementary (workload) traces

The FTA allows the following:

• the comparison and cross-validation of fault-tolerant model or algorithm across identical trace data sets
• the evaluation of the generality of a model or algorithm across different types of resources (e.g. in terms of reliability)
• the evaluation the generality of a failure trace, i.e., to determine whether measurements are biased to particular platform or middleware
• the incorporation of traces into fault simulators or emulators for model or algorithm evaluation
• the determination of which trace data set is most interesting or applicable for a given algorithm or model
• the integration of failure models with other types of models (such as workloads)