|8:40-9:00||Towards Ad Hoc Recovery For Soft Errors
Losada, Bautista-Gomez, Keller, Unsal
|9:00-9:20||Fault Tolerant Cholesky Factorization on GPUs
Loh, Saluja, Ramanathan
|Improving Application Resilience by Extending Error Correction with Contextual Information
Poulos, Wallace, Robey, Monroe, Job, Blanchard, Jones, DeBardeleben
|A Comprehensive Informative Metric for Analyzing HPC System Status using the LogSCAN Platform
Hui, Park, Engelmann
|Analyzing the Impact of System Reliability Events on Applications in the Titan Supercomputer
|Extending and Evaluating Fault-Tolerant Preconditioned Conjugate Gradient Methods
Pachajoa, Levonyak, Gansterer
|11:10-11:30||CPU Overheating Characterization in HPC Systems: a Case Study
Platini, Ropars, Pelletier, De Palma
|11:30-11:50||SaNSA - the Supercomputer and Node State Architecture
Agarwal, Greenberg, Blanchard, DeBardeleben
|Influence of A-Posteriori Subcell Limiting on Fault Frequency in Higher-Order DG Schemes
Reinarz, Gallard, Bader
Call for paper
Draft paper submission deadline：2018-09-10
Call for paper description
Authors are invited to submit original papers on the research and practice of fault-tolerance in extreme-scale distributed systems (primarily HPC systems, but including grid and cloud systems). Resilience and fault-tolerance remain a major concern for supercomputing and advances in this area are needed to allow applications to compute accurate (or within an acceptable error tolerance) answers in a timely and efficient manner in the presence of degradations or failures of platform components (both hardware and software).
Submissions are solicited in the following categories:
- Regular papers presenting innovative ideas improving the state of the art or discussing the issues seen on existing extreme-scale systems, including some form of analysis and evaluation.
- Extended abstracts proposing disruptive ideas and challenging assumptions in the field, including some form of preliminary results.
Extended abstracts will be evaluated separately and given shorter oral presentations.
Submissions shall be sent electronically and must conform to SC18 proceedings style. Regular papers should not exceed ten (10) pages including all text, appendices, figures, and references. Extended abstract papers should not exceed six (6) pages. Please note that we have only placed a limit on the maximum number of pages that a submission may contain. Papers that are clear, coherent, and complete (with the understanding that the submission may represent a work-in-progress) but are shorter than this maximum are encouraged.
Papers should be submitted to: https://submissions.supercomputing.org. A sample submission form is availablehere.
Our workshop has been accepted to have its proceedings published by IEEE TCHPC (and included in IEEE Xplore).
Authors are encouraged to include reproducibility artifacts as described on the conference website:
Inclusion of reproducibility artifacts is optional.
Submission of papers: August 30, 2018 September 10, 2018 (anywhere-on-earth)
Author notification: September 27, 2018 October 4, 2018 (anywhere-on-earth)
Camera ready papers: October 11, 2018
Workshop: Friday, November 16, 2018
Topics of submission
Topics include, but are not limited to:
- Failure data analysis and field studies
- Power, performance, resilience (PPR) assessments / tradeoffs
- Novel fault-tolerance techniques and implementations
- Emerging hardware and software technology for resilience
- Silent data corruption (SDC) detection / correction techniques
- Advances in reliability monitoring, analysis, and control of highly complex systems
- Failure prediction, error preemption, and recovery techniques
- Fault-tolerant programming models
- Models for software and hardware reliability
- Metrics and standards for measuring, improving, and enforcing effective fault-tolerance
- Scalable Byzantine fault-tolerance and security from single-fault and fail-silent violations
- Atmospheric evaluations relevant to HPC systems (terrestrial neutrons, temperature, voltage, etc.)
- Near-threshold-voltage implications and evaluations for reliability
- Benchmarks and experimental environments including fault injection
- Frameworks and APIs for fault-tolerance and fault management
Keita Teranishi – Sandia National Laboratories
John Daly – Laboratory for Physical Sciences