Zur NavigationLogin


G: Central soft matter simulation platform

The service project G is a central support project, which offers software development services for the science projects within TRR 146 and which is itself a research project that analyses and optimizes the usage of HPC in the context of soft-matter systems. Research challenges in project G include the development of load-balancing schemes for MD simulations, building an asynchronous checkpointing environment, which is able to improve reliability, and to allow applications to steer their IO-behavior and therefore to improve IO performance.

Scalable and fast heterogeneous molecular simulation with predictive parallelization schemes
Horacio V. Guzman, Christoph Junghans, Kurt Kremer, Torsten Stuehn
Physical Review E 96 (5), (2017);

MERCURY: a Transparent Guided I/O Framework for High Performance I/O Stacks
Giuseppe Congiu, Matthias Grawinkel, Federico Padua, James Morse, Tim Süß and André Brinkmann
in 25th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2017), IEEE Press (2017);

The performance gap between processors and I/O represents a serious scalability limitation for applications running on computing clusters. Parallel file systems often provide mechanisms that allow programmers to disclose their I/O pattern knowledge to the lower layers of the I/O stack through a hints API. This information can be used by the file system to boost the application performance. Unfortunately, programmers rarely make use of these features, missing the opportunity to exploit the full potential of the storage system. In this paper we propose MERCURY, a transparent guided I/O framework able to optimize file I/O patterns in scientific applications, allowing users to control the I/O behavior of applications without modifications. This is done by exploiting the hints API provided by the back-end file system to guide data prefetching. MERCURY efficiently converts numerous small read requests into a few larger requests. Furthermore, it increases the I/O bandwidth, reduces the number of I/O requests, and ultimately the application running time. Moreover, we also propose a Linux kernel modification that allows network file systems, specifically Lustre, to work with our guided I/O framework through the posix_fadvise interface.

Deduplication Potential of HPC Applications' Checkpoints
Jürgen Kaiser, Ramy Gad, Tim Süß, Federico Padua, Lars Nagel and André Brinkmann
in IEEE Int. Conf. on Cluster Computing (Cluster'16), Pages 413--422, IEEE Press (2016);

Analysis of the ECMWF Storage Landscape
Matthias Grawinkel, Lars Nagel, Markus Mäsker, Federico Padua, André Brinkmann, Lennart Sorth
in Proceedings of the 13th USENIX Conference on File and Storage Technologies {FAST} 2015, Santa Clara, CA, USA, Pages 15 - 27, Usenix (2015);

Optimizing scientific file {I/O} patterns using advice based knowledge
Giuseppe Congiu, Matthias Grawinkel, Federico Padua, James Morse, Tim Süß, André Brinkmann
in Proceedings of the International Conference on Cluster Computing (CLUSTER), Madrid, Spain, IEEE (2014);


This page uses cookies. By continuing to visit this site you agree to our use of cookies.

More info