arXiv subjects

Achim Streit

Publications and source records attributed to Achim Streit.

42 records · Page 3Linked to original sources

arXiv2016

A Distributed System for Storing and Processing Data from Earth-observing Satellites: System Design and Performance Evaluation of the Visualisation Tool

We present a distributed system for storage, processing, three-dimensional visualisation and basic analysis of data from Earth-observing satellites. The database and the server have been designed for high performance and scalability, whereas the client is highly portable thanks to having been designed as a HTML5- and WebGL-based Web application. The system is based on the so-called MEAN stack, a modern replacement for LAMP which has steadily been gaining traction among high-performance Web applications. We demonstrate the performance of the system from the perspective of an user operating the client.

cs.DC↗

arXiv2015

A horizontally-scalable multiprocessing platform based on Node.js

This paper presents a scalable web-based platform called Node Scala which allows to split and handle requests on a parallel distributed system according to pre-defined use cases. We applied this platform to a client application that visualizes climate data stored in a NoSQL database MongoDB. The design of Node Scala leads to efficient usage of available computing resources in addition to allowing the system to scale simply by adding new workers. Performance evaluation of Node Scala demonstrated a gain of up to 74 % compared to the state-of-the-art techniques.

cs.DC↗

arXiv2015

Challenges and Recommendations for Preparing HPC Applications for Exascale

While the HPC community is working towards the development of the first Exaflop computer (expected around 2020), after reaching the Petaflop milestone in 2008 still only few HPC applications are able to fully exploit the capabilities of Petaflop systems. In this paper we argue that efforts for preparing HPC applications for Exascale should start before such systems become available. We identify challenges that need to be addressed and recommend solutions in key areas of interest, including formal modeling, static analysis and optimization, runtime analysis and optimization, and autonomic computing. Furthermore, we outline a conceptual framework for porting HPC applications to future Exascale computing systems and propose steps for its implementation.

cs.DC↗

arXiv2014

Architecture, implementation and parallelization of the software to search for periodic gravitational wave signals

The parallelization, design and scalability of the \sky code to search for periodic gravitational waves from rotating neutron stars is discussed. The code is based on an efficient implementation of the F-statistic using the Fast Fourier Transform algorithm. To perform an analysis of data from the advanced LIGO and Virgo gravitational wave detectors' network, which will start operating in 2015, hundreds of millions of CPU hours will be required - the code utilizing the potential of massively parallel supercomputers is therefore mandatory. We have parallelized the code using the Message Passing Interface standard, implemented a mechanism for combining the searches at different sky-positions and frequency bands into one extremely scalable program. The parallel I/O interface is used to escape bottlenecks, when writing the generated data into file system. This allowed to develop a highly scalable computation code, which would enable the data analysis at large scales on acceptable time scales. Benchmarking of the code on a Cray XE6 system was performed to show efficiency of our parallelization concept and to demonstrate scaling up to 50 thousand cores in parallel.

gr-qc↗

arXiv2014

Energy-Aware Cloud Management through Progressive SLA Specification

Novel energy-aware cloud management methods dynamically reallocate computation across geographically distributed data centers to leverage regional electricity price and temperature differences. As a result, a managed VM may suffer occasional downtimes. Current cloud providers only offer high availability VMs, without enough flexibility to apply such energy-aware management. In this paper we show how to analyse past traces of dynamic cloud management actions based on electricity prices and temperatures to estimate VM availability and price values. We propose a novel SLA specification approach for offering VMs with different availability and price values guaranteed over multiple SLAs to enable flexible energy-aware cloud management. We determine the optimal number of such SLAs as well as their availability and price guaranteed values. We evaluate our approach in a user SLA selection simulation using Wikipedia and Grid'5000 workloads. The results show higher customer conversion and 39% average energy savings per VM.

cs.DC↗

arXiv2012

Data Life Cycle Labs, A New Concept to Support Data-Intensive Science

In many sciences the increasing amounts of data are reaching the limit of established data handling and processing. With four large research centers of the German Helmholtz association the Large Scale Data Management and Analysis (LSDMA) project supports an initial set of scientific projects, initiatives and instruments to organize and efficiently analyze the increasing amount of data produced in modern science. LSDMA bridges the gap between data production and data analysis using a novel approach by combining specific community support and generic, cross community development. In the Data Life Cycle Labs (DLCL) experts from the data domain work closely with scientific groups of selected research domains in joint R&D where community-specific data life cycles are iteratively optimized, data and meta-data formats are defined and standardized, simple access and use is established as well as data and scientific insights are preserved in long-term and open accessible archives.

cs.DL↗