525 research outputs found
Data for the publication "Ozone concentration versus Temperature: Atmospheric aging of soot particles"
No full textThe repository contains the data for the paper:
Friebel, F and Mensah, A. A. Ozone concentration versus Temperature: Atmospheric aging of soot particles, submitted to Langmuir 2019, in review (accepted September 2019)</p
Automatic Creation of High-Bandwidth Memory Architectures from Domain-Specific Languages: The Case of Computational Fluid Dynamics
No full textNumerical simulations can help solve complex problems. Most of these
algorithms are massively parallel and thus good candidates for FPGA
acceleration thanks to spatial parallelism. Modern FPGA devices can leverage
high-bandwidth memory technologies, but when applications are memory-bound
designers must craft advanced communication and memory architectures for
efficient data movement and on-chip storage. This development process requires
hardware design skills that are uncommon in domain-specific experts. In this
paper, we propose an automated tool flow from a domain-specific language (DSL)
for tensor expressions to generate massively-parallel accelerators on
HBM-equipped FPGAs. Designers can use this flow to integrate and evaluate
various compiler or hardware optimizations. We use computational fluid dynamics
(CFD) as a paradigmatic example. Our flow starts from the high-level
specification of tensor operations and combines an MLIR-based compiler with an
in-house hardware generation flow to generate systems with parallel
accelerators and a specialized memory architecture that moves data efficiently,
aiming at fully exploiting the available CPU-FPGA bandwidth. We simulated
applications with millions of elements, achieving up to 103 GFLOPS with one
compute unit and custom precision when targeting a Xilinx Alveo U280. Our FPGA
implementation is up to 25x more energy-efficient than expert-crafted Intel CPU
implementations.Comment: Accepted for publication in ACM Transactions on Reconfigurable
Technology and Systems (TRETS
From Domain-Specific Languages to Memory-Optimized Accelerators for Fluid Dynamics
No full textMany applications are increasingly requiring numerical simulations for solving complex problems. Most of these numerical algorithms are massively parallel and often implemented on parallel high-performance computers. However, classic CPU-based platforms suffer due to the demand for higher resolutions and the exponential growth of data. FPGAs offer a powerful and flexible alternative that can host accelerators to complement such platforms. Developing such application-specific accelerators is still challenging because it is hard to provide efficient code for hardware synthesis. In this paper, we study the challenges of porting a numerical simulation kernel onto FPGA.
We propose an automated tool flow from a domain-specific language (DSL) to generate accelerators for computational fluid dynamics on FPGA. Our DSL-based flow simplifies the exploration of parameters and constraints such as on-chip memory usage.
We also propose a decoupled optimization of memory and logic resources, which allows us to better use the limited FPGA resources.
In our preliminary evaluation, this enabled doubling the number of parallel kernels, increasing the accelerator speedup versus ARM execution from 7 to 12 times
Data for the publication "Impact of Isolated Atmospheric Aging processes on the Cloud Condensation Nucleiactivation of Soot Particles"
No full textThe repository contains the data for the paper:
Friebel, F.; Lobo, P.; Neubauer, D.; Lohmann, U.; Drossaart van Dusseldorp, S.; Mühlhofer, E.; Mensah, A. A. Impact of Isolated Atmospheric Aging Processes on the Cloud Condensation Nuclei-Activation of Soot Particles. Atmos. Chem. Phys. Discuss. 2019, No. June, 1–28. https://doi.org/10.5194/acp-2019-504.
Note that the scripts to plot this data are to be found in the accompanying package (http://dx.doi.org/10.5281/zenodo.3452036)</p
Human Smuggling
No full textDespite its importance in global illegal migration, there is little, and mostly theoretical research on human smuggling. We suggest an analytical framework to understand the micro structure of the human smuggling market. Migrants interact with smuggling and financing intermediaries; these may or may not be integrated with each other, and with the migrants' employers. Policies of receiving countries (border controls, employer sanctions, deportation policies, sales of visa) affect the interactions in the smuggling market, and, hence, migration flows. We review the theoretical work, point to the scarce empirical evidence, and identify challenges for future theoretical, empirical work and policy advice.illegal migration, trafficking
Generative Transcriptions: An Opera of the Self
No full textThe techniques which I developed for my work between 2009-12 are a consequence of my own unique mix of synaesthetic sensibilities. I have sought to investigate these innate impulses in order to formulate a deeper understanding of my compositional process and define technical strategies that I have termed generative transcriptions.
In this thesis I focus on CANTO MORPH, ‘an opera of the self’, a project in multiple parts exploring processes of embodiment, association and dissociation. It is realised through the media of recorded improvisation, a 45-minute scored composition, a virtual architecture, a media installation, a video installation and a sculptural libretto.
This project as ‘an opera of the self’ is located in an experiential process comprising several stages:
1/ I perform a series of improvisations in a trance-like state using various instruments and my own voice to create multi-layered recordings – trance improvisation mix.
2/ These recordings are subjected to analytical transcription which I understand as a process of reverse engineering akin to the process that architects employ whereby complex 3-dimensional structures of buildings are analysed and explored diagrammatically in order to discover what kind of essence might be contained in the blueprint of its built form. I cite Xenakis’ Philips Pavilion and the dual nature of composition/architecture that is embodied within this design. This process inspires my own methodology, where mimetic scripts are created from the trance improvised mix and then refined and further manipulated into a score that can be performed.
3/ The improvisation-analysis-composition process which can broadly be conceived as an ‘embodiment of self’ is subjected to another stage in which I create a new body – a re-embodiment in which to site the work architecturally.
In exploring the multi-modal approach that is represented by the CANTO MORPH, I have underpinned a novel and multi-facetted way to approach a compositional practice, where through synaesthetic empowerment, a myriad of forces enable an emergent compositional method of generative transcriptions
Promoting the systematic use of real-world data and real-world evidence for digital health technologies across Europe: a consensus framework
No full textDespite the acceleration in the use of digital health technologies across different aspects of the healthcare system, the full potential of real-world data (RWD) and real-world evidence (RWE) arising from the technologies is not being utilised in decision-making. We examine current national efforts and future opportunities to systematically use RWD and RWE in decision-making in five countries (Estonia, Finland, Germany, Italy and the United Kingdom), and then develop a framework for promotion of the systematic use of RWD and RWE. A review assesses current national efforts, complemented with a three-round consensus-building exercise among an international group of experts (n1 = 44, n2 = 24, n3 = 24) to derive key principles. We find that Estonia and Finland have invested and developed digital health-related policies for several years; Germany and Italy are the more recent arrivals, while the United Kingdom falls somewhere in the middle. Opportunities to promote the systematic use of RWD and RWE were identified for each country. Eight building blocks principles were agreed through consensus, relating to policy scope, institutional role and data collection. Promoting post-market surveillance and digital health technology vigilance ought to rely on clarity in scope and data collection with consensus reached on eight principles to leverage RWD and RWE
Macromolecular Specifity and Biological Memory. Herausgegeben von F. O. Schmitt, 119 Seiten. Preis $ 3.– The M. I. T. Press, Cambridge, Mass. 1962
No full text- …
