. .

Projects

Scalasca – Scalable Performance Analysis of Large-Scale Parallel Applications

Scalasca is a long-term umbrella project to which many of our activities contribute. It is an open-source toolset developed in cooperation with the Jόlich Supercomputing Centre that can be used to analyze the performance behavior of parallel applications and to identify opportunities for optimization. It has been specifically designed for use on large-scale systems, but is also well-suited for small- and medium-scale HPC platforms. Scalasca integrates runtime summaries with in-depth studies of concurrent behavior via event tracing. A distinctive feature is its ability to identify wait states that occur, for example, as a result of unevenly distributed workloads. More

Catwalk – A Quick Development Path for Performance Models

The objective of this DFG-funded project is to provide a flexible set of tools to support key activities of the performance modeling process, making this powerful methodology accessible to a wider audience of HPC application developers. The tool suite will be used to study and help improve the scalability of applications from life sciences, fluid dynamics, and particle physics. The project is part of the DFG Priority Programme 1648 Software for Exascale Computing (SPPEXA). More

EESI 2 – European Exascale Software Initiative 2

The constant growth of generated data - Big Data - and computing capabilities of extreme systems lead to a new generation of computers composed of millions of heterogeneous cores and expected to provide Exaflop performances around 2020. The successful first European Exascale Software Initiative (EESI 1) federated the European community, built a preliminary European cartography, vision and roadmap and served as the European voice on the international level. In the follow-up project, EESI 2, we go one step further towards implementation by establishing a European structure to gather the European community and by providing periodically cartography and roadmaps and dynamic synthesis and recommendations. More

DEEP – Dynamical ExaScale Entry Platform

The goal of this EU project is to develop a novel, exascale-enabling supercomputing architecture with a matching softeware stack and a set of optimized grand-challenge simulation applications. DEEP takes the concept of compute acceleration to a new level: instead of adding accelerator cards to cluster nodes, an accelerator cluster, called booster, will complement a conventional HPC system and increase its compute performance. GRS contributes in the areas of middleware, resource management, and performance tools. More

LMAC – Performance Dynamics of Massively Parallel Codes

The time-dependent behavior of parallel simulation codes is often irregular, making the understanding of performance dynamics an essential prerequisite for program optimization. To better support developers in optimizing the dynamic behavior of their codes, this BMBF-funded project extends the performance analysis tools Vampir, Scalasca, and Periscope with new functionality to automatically examine time-dependent performance phenomena. In addition, the University of Oregon, associated partner, complements the project with corresponding extensions to the performance tool TAU. More

HOPSA – Holistic Performance System Analysis

The EU project HOPSA in the framework of the EU-Russia Coordinated Call sets out to create an integrated diagnostic infrastructure for combined application and system tuning – with the former being under EU and the latter being under Russian responsibility. Starting from system-wide basic performance screening of individual jobs, an automated workflow will route findings on po-tential bottlenecks either to system administrators or application developers with recommendations on how to identify their root cause using more powerful diagnostic tools. More

G8 ECS – Enabling Climate Simulation at Extreme Scale

The objective of the G8 ECS project, which is locally funded by the DFG in the framework of the G8 Research Councils Initiative on Multilateral Research (Interdisciplinary Program on Application Software towards Exascale Computing), is to investigate how to run efficiently climate simulations on future Exascale systems. It focuses on three main topics: (i) how to complete simulations with correct results despite frequent system failures, (ii) how to exploit hierarchical computers with hardware accelerators close to their peak performance and (iii) how to run efficient simulations with 1 billion threads. More

PRIMA – Performance Refactoring of Instrumentation, Measurement, and Analysis Technologies for Petascale Computing

This DOE-funded project pursues the goal of reengineering core components of the two performance-analysis systems TAU, developed by the University of Oregon, and Scalasca for evolution to petascale and beyond. Building on a long history of interaction between the two projects, the two key activities are 1) refactoring certain TAU and Scalasca components for core code sharing, and 2) integrating their functionality more effectively through data interfaces, formats, and utilities. More

VI-HPS – Virtual Institute High Productivity Supercomputing

Funded by the Helmholtz Association, the mission of the Virtual Institute - High Productivity Supercomputing is to improve the quality and accelerate the development process of complex simulation programs in science and engineering that are being designed for the most advanced parallel computer systems. For this purpose, eleven partners in Europe and the US are developing and integrating state-of-the-art programming tools for high-performance computing that assist domain scientists in diagnosing programming errors and optimizing the performance of their applications. Besides the purely technical development of programming tools, the virtual institute also offers training workshops with guided hands-on training in the effective use of the tools. More

AICES – Aachen Institute for Advanced Study in Computational Engineering Science

The Graduate School Aachen Institute for Advanced Study in Computational Engineering Science (AICES) is a new doctoral program established at RWTH Aachen University in November 2006 under the auspices of the Excellence Initiative of the German Federal Government and the governments of the federal states as part of the line for funding graduate schools. The program sets out to advance computational engineering in three critical areas of synthesis: model identification and discovery supported by model-based experimentation, understanding scale interaction and scale integration, and optimal design and operation of engineered systems. More

Past Projects