Call For Papers [PDF] [DOC]

Since early days of computing systems, researchers and engineers have recognized the crucial role of Field Failure Data Analysis (FFDA) in characterizing the failure behavior of systems and guiding the design of error detection and recovery mechanisms. The information on spontaneous occurrences of systems/applications failures is usually collected in a variety of error logs. Analysis of this data can provide useful insight into understanding of error propagation patterns, enables identification of dependability bottlenecks, and quantification of metrics such as reliability, availability, or sensitivity to malicious attacks. Despite of the many studies analyzing failure data from variety of computing systems, a number of research and practical questions remain unanswered. For instance, it is well-understood that the quality of the analysis heavily depends on the quality of the available field failure data (i.e., system logs or failure reports). However, there are no commonly accepted criteria and/or methodologies to define how (what sources to use) to gather high quality data and how to process the data to obtain the meaningful unbiased results. Also, it is not clear whether traditional FFDA techniques are suitable to conduct the dependability evaluation of current and future generation systems, e.g., autonomic and/or ubiquitous systems, embedded systems, and mashup web applications.

Goals : This workshop aims to foster lively discussion and advance the state-of-the art in filed failure data analysis of current and future systems. The event provides an open forum to industry practitioners and academia to share experience and ideas on open issues and future trends in analyzing and using the field failure data.

The major areas of interest include, but are not limited to, the following:

  • Methods, tools, and infrastructures for field data gathering and management
  • Data formats and data archiving
  • Methodologies and tools for failure data processing, e.g., efficient algorithms for data filtering and correlation analysis
  • Failure data driven design of dependable systems and applications
  • Challenges in collecting and analyzing failure data from emerging systems/applications, e.g., smart handheld devices or sensor networks
  • Dependability evaluation using field data

Workshop Committee:

  • Chair: Domenico Cotroneo, University of Napoli, Italy
  • Co-Chair: Zbigniew Kalbarczyk, University of Illinois at Urbana-Champaign
  • Bianca Schroeder, Carnegie Mellon University, USA
  • Brendan Murphy, Microsoft Research,UK
  • Marcello Cinque, Univ. of Naples Federico II
  • Marco Vieira, University of Coimbra, Portugal
  • Miroslaw Malek Humboldt University Berlin, Germany
  • Paolo Lollini, University of Florence, Italy
  • Brendan Murphy, Microsoft Research,UK

Submission Guidelines:

Please see the submission page for information.

Workshop Program: (pdf version)

List of Accepted Papers

  1. M. Cinque, R. Natella, A.Pecchia, S. Russo, Improving FFDA of Web Servers through a Rule-Based Logging Approach, University of Naples Federico II, Italy
  2. Cobra Rahmani, Harvey Siy, Azad Azadmanesh, An Experimental Analysis of Open Source Software Reliability, University of Nebraska-Omaha, U.S.
  3. Ambili Thottam Parameswaran, Mohammad Iftekhar Husain and Shambhu Upadhyaya. Is RSSI a reliable parameter in sensor localization algorithms – an experimental study, State University of New York at Buffalo, U.S.