UCER 2016 - Ultrascale Computing for Early Researchers: UCER 2016

The aim of this workshop is to give the opportunity to early researchers (PhD students or recent PhD graduates) to show their work related to Ultrascale Computing. Although a future technology, currently, many systems are being designed with the goal of being used in ultrascale systems. Many different subtopics are related in the exploration of system software and applications for enabling a sustainable development of future high-scale computing platforms. The tasks involved range from the analysis of the current state-of-the-art on sustainability in large-scale systems to the proposition of new tools that aim to improve computations on these systems. The topics addressed are, among others, HPC, distributed systems, and big data communities in cross cutting aspects like programmability, scalability, resilience, energy efficiency, and data management. To get the goal of ultrascale computation it is needed to explore new programming paradigms, runtimes, and middlewares to increase the productivity, scalability, and reliability of parallel and distributed programming. At the same time, the new magnitude of data and computations brings up as inevitability consequence the probability of failure, so any advance on resilient schedulers that handle errors reactive or proactive, monitoring and assessment of failures, and malleable applications that can adapt their resource usage at runtime are welcome. Other major challenges involved are the restructuring the Input/Output (I/O) stack, the advancing predictive and adaptive data management, and the concern about huge energy consumption as one of the major limitations. This topic also includes identifying applications, high-level algorithms, and services amenable to ultrascale systems and investigating the redesign and reprogramming efforts needed for applications to efficiently exploit ultrascale platforms while providing sustainability.
UCER 2016 welcomes original submissions that have not been published and that are not under review by another conference or journal. To respect the workshop's aim, at least one author has to be a PhD student or recent PhD graduate. All submitted papers will be peer-reviewed by UCER's technical program committee. All submissions will be evaluated on significance, presentation, and interest to the workshop attendees.
The program committee cordially invites any novel research ideas in the following (but not limited to) topics:
Parallel and distributed systems for ultrascale computing, including, frameworks, engines or programming models.
GPU and Heterogeneous computing
Data management methods and techniques for ultrascale computing.
Fault tolerance techniques.
Energy efficiency: monitoring, evaluation, modelling.
Load balancing and scheduling
Applications suitable for ultrascale computing.