HotDep 2005

First Workshop on Hot Topics
in System Dependability

June 30, 2005
Yokohama, Japan

held in conjunction with the
International Conference on Dependable Systems and Networks (DSN-2005)

HotDep '05 is aimed at identifying cutting-edge research ideas spanning the domains of systems and fault tolerance/reliability. The workshop focuses on critical components of the infrastructures that touch our everyday lives: operating systems, networking, security, wide-area and enterprise-scale distributed systems, mobile computing, etc. The workshop proceedings consist of position papers from both academic researchers and industry practitioners, to achieve a mix of long-range research vision and technology ideas anchored in immediate reality.

Preliminary Program
8:30 Introduction
8:35 Distributed Systems

The Virtue of Dependent Failures in Multi-Site Systems [PDF]
Flavio P. Junqueira, Keith Marzullo (U.C. San Diego)

A Root-Cause Localization Model for Large-Scale Systems [PDF]
Emre Kiciman (Stanford University), Lakshminarayanan Subramanian (U.C. Berkeley)

The Role of Accountability in Dependable Distributed Systems [PDF]
Aydan R. Yumerefendi, Jeffrey S. Chase (Duke University)

Q&A Mini-panel: Flavio Junqueira, Emre Kiciman, Aydan Yumerefendi
10:00 Coffee break
10:30 Dependable Services

Trusted Virtual Domains: Toward Secure Distributed Services [PDF]
John Linwood Griffin, Trent Jaeger, Ronald Perez, Reiner Sailer, Leendert van Doorn, Ramón Cáceres (IBM Research)

Computational Risk Management for Building Highly Reliable Network Services [PDF]
Brent N. Chun, Philip Buonadonna (Intel Research, Berkeley), Chaki Ng (Harvard University)

On the Challenge of Delivering High-Performance, Dependable, Model-Checked Internet Servers [PDF]
Anil Madhavapeddy (University of Cambridge), David Scott (Fraser Research)

Q&A Mini-panel: John Linwood Griffin, Anil Madhavapeddy, Chaki Ng
12:00 Lunch break
13:00 Challenge Your Assumptions

Why Traditional Storage Systems Don't Help Us Save Stuff Forever [PDF]
Mary Baker, Kim Keeton (HP Labs), Sean Martin (British Library, UK)

What Dependability for Networks of Mobile Sensors ? [PDF]
Carole Delporte-Gallet, Hugues Fauconnier (University of Paris), Rachid Guerraoui (EPFL)

Application Communities: Using Monoculture for Dependability [PDF]
Michael E. Locasto, Stelios Sidiroglou, Angelos D. Keromytis (Columbia University)

TACID Transactions [PDF]
Marco Vieira (University of Coimbra), António C. Costa (University of Lisbon), Henrique Madeira (University of Coimbra)

Q&A Mini-panel: Mary Baker, António Costa, Rachid Guerraoui, Angelos Keromytis
15:00 End of workshop
Papers with Posters

Handling Cascading Failures: The Case for Topology-Aware Fault Tolerance [PDF]
Soila Pertet, Priya Narasimhan (Carnegie Mellon University)

The Case for an Internet Health Monitoring System [PDF]
Matthew Caesar, Lakshminarayanan Subramanian, Randy H. Katz (U.C. Berkeley)

Rewriting "The Turtle and the Hare": Sleeping to Get There Faster [PDF]
José Pereira, Rui Oliveira (University of Minho)

Time-varying Management of Data Storage [PDF]
Ranjita Bhagwan, Fred Douglis, Kirsten Hildrum, Jeffrey O. Kephart, William E. Walsh (IBM T.J. Watson Research Center)

Managing Self-Inflicted Nondeterminism [PDF]
Dmitrii Zagorodnov (University of Tromso), Keith Marzullo (U.C. San Diego)


Program Co-Chairs

George Candea, Stanford University
David Oppenheimer University of California, Berkeley
Program Committee
Lorenzo Alvisi University of Texas, Austin
Christian Cachin IBM Research, Zurich
Valérie Issarny INRIA / Institut National de Recherche en Informatique et en Automatique
Ravishankar Iyer University of Illinois, Urbana-Champaign
Kimberly Keeton HP Labs, Palo Alto
Angelos Keromytis Columbia University
Keith Marzullo University of California, San Diego
Priya Narasimhan Carnegie Mellon University
David Patterson University of California, Berkeley
Martin Rinard Massachusetts Institute of Technology
Daniel Siewiorek Carnegie Mellon University
Amin Vahdat University of California, San Diego
Werner Vogels
  • IEEE Computer Society Technical Committee on Fault-Tolerant Computing
  • IFIP Workgroup 10.4 on Dependable Computing and Fault Tolerance
  • IEICE Technical Group on Dependable Computing

Original Call for Papers

Authors are invited to submit position papers to the First Workshop on Hot Topics in System Dependability. The goal of this workshop is to identify cutting-edge research ideas spanning the domains of fault tolerance and reliability (e.g., as reflected at conferences such as DSN and ISSRE) and systems (e.g., as reflected at conferences such as OSDI and SOSP). We are today at a historic confluence of interests between the two communities, with the former bringing decades of fault tolerance research into the domain of general purpose systems, and the latter taking an earnest interest in dependability. It is our hope that providing mutual sounding boards for plans for future research will benefit both communities and bring them closer together.

The workshop will focus on critical components of the infrastructures touching our everyday lives: operating systems, networking, security, wide-area and enterprise-scale distributed systems, mobile computing, compilers, and language design. We seek participation and contributions from both academic researchers and industry practitioners, to achieve a mix of long-range research vision and technology ideas anchored in immediate reality.

Position papers (max. 5 pages) should preferably fall into one of the following categories:

  • describe a novel approach to an old problem, that promises to influence future research
  • debunk an old, entrenched perspective on dependability
  • articulate a brand new perspective on existing problems in dependability
  • describe a new problem (and possible solution) that must be addressed by the dependable systems research community

The program committee will favor papers that are likely to generate healthy debate at the workshop. Ideas do not have to be 100% fleshed out and/or entirely backed up by quantitative measurements, but must provide credible evidence that they are feasible and must be accompanied by a compelling motivation. Possible topics include, but are not limited to:

  • automated failure management, that enables systems to adapt on the fly to exceptional conditions
  • techniques for better detection, diagnosis, or recovery from failures
  • forensic tools for administrators and programmers to use after a failure or attack
  • techniques and metrics for quantifying aspects of dependability in specific domains (e.g., measuring the security of a Web service)
  • tools/concepts/techniques for optimizing tradeoffs among availability, performance, correctness, and security
  • novel uses of technologies not originally intended for dependability (e.g., using virtual machines to enhance dependability)

All accepted papers will be available online prior to the workshop and will be published in the supplement to the DSN proceedings. A summary of the workshop will be published in the DSN main proceedings.

The workshop's homepage is at


Submission deadline March 1, 2005 (Firm deadline, absolutely no extensions)
Acceptance notification     March 30, 2005
Camera-ready deadline April 28, 2005

Submitting a paper

Position papers must be received by 11:59 PM Pacific Standard Time on March 1, 2005. Submissions should contain no more than 5 two-column pages, including all figures and references, single-space, using 11-point font, and 1-inch margins. To submit your paper, please email it in PDF or PostScript format to; receipt will be confirmed within 24 hours. Author names and affiliations should appear on the title page.

Comments/suggestions to George Candea. Last updated April 14, 2004.