scholarly journals VECTOR CLOCK TRACING AND MODEL BASED PARTITIONING FOR DISTRIBUTED EMBEDDED SYSTEMS

2014 ◽  
pp. 324-332
Author(s):  
Robert Hoettger ◽  
Burkhard Igel ◽  
Erik Kamsties
Keyword(s):  
Performance Optimization ◽  
High Performance ◽  
Distributed Embedded Systems ◽  
Control Engineering ◽  
Fine Grained ◽  
Model Based ◽  
Optimal Load ◽  
Vector Clock ◽  
Fine Grained Structure ◽  
High Degree

Tracking, partitioning and tracing in modern dynamic high performance computing systems are three of the most innovative and important development aspects for performance optimization purposes and state-of-the-art advanced quality. This paper discusses these three aspects with respect to distributed systems and proposes new mechanisms for an advanced utilization of software in this domain. We present a specific tracking mechanism via vector clocks for model and code partitioning purposes and the determination of causality relations. Further, a tracing approach for an effective analysis and thereby utilization of code and the corresponding architecture is introduced. The combination of both approaches leads to a high degree of parallelism and a fine-grained structure of execution units, that further traced, supports a precise analysis of synchronous and asynchronous system’s behavior as well as an optimal load balancing. The mechanisms are introduced with respect to a model based control engineering tool and event diagrams.

scholarly journals Development of actions on optimization of chemical composition of carburized steel grade 16MnCrS5 to eliminate the causes of the growth of large austenite grains

2019 ◽  
pp. 49-56 ◽  
Author(s):  
I. A. Kovaleva ◽  
I. A. Ovchinnikova ◽  
S. V. Stefanovitch
Keyword(s):  
Grain Size ◽  
Chemical Composition ◽  
High Performance ◽  
Internal Combustion Engines ◽  
Cutting Tools ◽  
Steel Grade ◽  
Fine Grained ◽  
Carburized Steel ◽  
Fine Grained Structure ◽  
Steel Grades

The problem of friction, wear of machine parts and cutting tools, the need to increase the life of machines makes the task of creating new high-performance, energy-saving technologies of hardening one of the main in engineering. In mechanical engineering, the problem of improving the physical, mechanical and operational properties in thin surface layers (~10 microns) of such parts as shafts, gears, measuring tools, drills, cylinders of internal combustion engines, etc. is important. Currently, these tasks are often solved by applying reinforcing coatings. Cementation is economical. The technology of cemented steels smelting, which is currently used by metallurgical plants, does not ensure stable production of fine-grained structure in products. At the same time, the capacity of machines and units are growing, and to obtain parts that can withstand higher loads while maintaining their overall dimensions, new steel grades are needed. The complexity of the configuration of gears with a thin tooth and a massive sleeve, and the need for minimal warping make heat treatment as difficult and responsible as a complex tool, such as shaped cutters. Therefore, for the manufacture of gears you want to assign steel with small hereditary grain size. In this article we will focus on the development and production of cemented steel in the conditions of OJSC «BSW – Management Company of Holding «BMC», in particular, the steel brand 16MnCrS5 commissioned by the European manufacturer of gearboxes and motors. To determine the grain size of the metal must be subjected to special types of processing for the manifestation of certain characteristics. These results allow us to draw conclusions about the need to adjust the chemical composition of steel grade 16MnCrS5 by introducing a system of modifying elements, which will reduce the tendency of steel to overheating, therefore, reduce the size of austenitic grains. On the basis of the revealed regularities on the influence of carbide-forming elements, a further system of steel modification is determined, which includes a complex of elements V, Nb, Ti.

Numerical Simulation of Cohesion Influence of Textile Reinforcement on Bending Performance of Plates Prepared from High Performance Concrete (HPC)

2015 ◽  
Vol 1106 ◽  
pp. 69-72 ◽  
Author(s):  
Tomáš Vlach ◽  
Alexandru Chira ◽  
Lenka Laiblová ◽  
Ctislav Fiala ◽  
Magdaléna Novotná ◽  
...  
Keyword(s):  
High Performance ◽  
High Performance Concrete ◽  
Concrete Cover ◽  
Textile Reinforced Concrete ◽  
Significant Saving ◽  
Bending Performance ◽  
Fine Grained ◽  
Technical Textiles ◽  
Fine Grained Structure ◽  
New Type

Demand for very thin concrete elements, which can’t be reinforced with usually used steel reinforcement, gave rise to a new type of non-traditional reinforcement with technical textiles in matrix of epoxy resin. This type of reinforcement together with concrete is called textile reinforced concrete (TRC). Composite reinforcement is very chemically resistant, so the concrete cover is proposed to regard the durability. It allows a significant saving of concrete and design of thinner elements. For TRC structures is used high performance concrete (HPC) with its fine grained structure and high compressive strength. Textile reinforcement and TRC in general are developed at the Faculty of Civil Engineering and the Klokner Institute, CTU in Prague.

Spatial Variability of Contaminant Transport in a Fractured Till, Avedøre Denmark

1999 ◽  
Vol 30 (4-5) ◽  
pp. 333-360 ◽  
Author(s):  
Larry McKay ◽  
Johnny Fredericia ◽  
Melissa Lenczewski ◽  
Jørn Morthorst ◽  
Knud Erik S. Klint
Keyword(s):  
Hydraulic Conductivity ◽  
Spatial Variability ◽  
Field Experiment ◽  
Contaminant Transport ◽  
Tracer Test ◽  
Conductivity Measurements ◽  
Fine Grained ◽  
Downward Migration ◽  
Rapid Transport ◽  
High Degree

A field experiment shows that rapid downward migration of solutes and microorganisms can occur in a fractured till. A solute tracer, chloride, and a bacteriophage tracer, PRD-1, were added to groundwater and allowed to infiltrate downwards over a 4 × 4 m area. Chloride was detected in horizontal filters at 2.0 m depth within 3-40 days of the start of the tracer test, and PRD-1 was detected in the same filters within 0.27 - 27 days. At 2.8 m depth chloride appeared in all the filters, but PRD-1 appeared in only about one-third of the filters. At 4.0 m depth chloride appeared in about one-third of the filters and trace amounts of PRD-1 were detected in only 2 of the 36 filters. Transport rates and peak tracer concentrations decreased with depth, but at each depth there was a high degree of variability. The transport data is generally consistent with expectations based on hydraulic conductivity measurements and on the observed density of fractures and biopores, both of which decrease with depth. Transport of chloride was apparently retarded by diffusion into the fine-grained matrix between fractures, but the rapid transport of PRD-1, with little dispersion, indicates that it was transported mainly through the fractures.

scholarly journals MODES: model-based optimization on distributed embedded systems

2021 ◽  
Author(s):  
Junjie Shi ◽  
Jiang Bian ◽  
Jakob Richter ◽  
Kuan-Hsun Chen ◽  
Jörg Rahnenführer ◽  
...  
Keyword(s):  
Machine Learning ◽  
Embedded Systems ◽  
Learning Model ◽  
Black Box ◽  
Distributed Embedded Systems ◽  
Data Set ◽  
Individual Model ◽  
Model Based ◽  
Machine Learning Model ◽  
Distributed Machine Learning

AbstractThe predictive performance of a machine learning model highly depends on the corresponding hyper-parameter setting. Hence, hyper-parameter tuning is often indispensable. Normally such tuning requires the dedicated machine learning model to be trained and evaluated on centralized data to obtain a performance estimate. However, in a distributed machine learning scenario, it is not always possible to collect all the data from all nodes due to privacy concerns or storage limitations. Moreover, if data has to be transferred through low bandwidth connections it reduces the time available for tuning. Model-Based Optimization (MBO) is one state-of-the-art method for tuning hyper-parameters but the application on distributed machine learning models or federated learning lacks research. This work proposes a framework $$\textit{MODES}$$ MODES that allows to deploy MBO on resource-constrained distributed embedded systems. Each node trains an individual model based on its local data. The goal is to optimize the combined prediction accuracy. The presented framework offers two optimization modes: (1) $$\textit{MODES}$$ MODES -B considers the whole ensemble as a single black box and optimizes the hyper-parameters of each individual model jointly, and (2) $$\textit{MODES}$$ MODES -I considers all models as clones of the same black box which allows it to efficiently parallelize the optimization in a distributed setting. We evaluate $$\textit{MODES}$$ MODES by conducting experiments on the optimization for the hyper-parameters of a random forest and a multi-layer perceptron. The experimental results demonstrate that, with an improvement in terms of mean accuracy ($$\textit{MODES}$$ MODES -B), run-time efficiency ($$\textit{MODES}$$ MODES -I), and statistical stability for both modes, $$\textit{MODES}$$ MODES outperforms the baseline, i.e., carry out tuning with MBO on each node individually with its local sub-data set.

High performance parallel-DSP computing in model-based spectral estimation

1999 ◽  
Vol 23 (6) ◽  
pp. 337-344 ◽  
Author(s):  
J.S. González ◽  
D.F. Garcı́a Nocetti ◽  
M.G. Ruano
Keyword(s):  
High Performance ◽  
Spectral Estimation ◽  
Model Based

scholarly journals Core-Level Modeling and Frequency Prediction for DSP Applications on FPGAs

2015 ◽  
Vol 2015 ◽  
pp. 1-20
Author(s):  
Gongyu Wang ◽  
Greg Stitt ◽  
Herman Lam ◽  
Alan George
Keyword(s):  
High Performance ◽  
Design Space Exploration ◽  
Design Space ◽  
Space Exploration ◽  
Core Level ◽  
Prediction Methods ◽  
Clock Frequency ◽  
Worst Case ◽  
Model Based ◽  
Dsp Applications

Field-programmable gate arrays (FPGAs) provide a promising technology that can improve performance of many high-performance computing and embedded applications. However, unlike software design tools, the relatively immature state of FPGA tools significantly limits productivity and consequently prevents widespread adoption of the technology. For example, the lengthy design-translate-execute (DTE) process often must be iterated to meet the application requirements. Previous works have enabled model-based, design-space exploration to reduce DTE iterations but are limited by a lack of accurate model-based prediction of key design parameters, the most important of which is clock frequency. In this paper, we present a core-level modeling and design (CMD) methodology that enables modeling of FPGA applications at an abstract level and yet produces accurate predictions of parameters such as clock frequency, resource utilization (i.e., area), and latency. We evaluate CMD’s prediction methods using several high-performance DSP applications on various families of FPGAs and show an average clock-frequency prediction error of 3.6%, with a worst-case error of 20.4%, compared to the best of existing high-level prediction methods, 13.9% average error with 48.2% worst-case error. We also demonstrate how such prediction enables accurate design-space exploration without coding in a hardware-description language (HDL), significantly reducing the total design time.

Sign in / Sign up

Export Citation Format

Share Document