Browsing by Author "Tkachev, Gleb"

Now showing 1 - 3 of 3

Open Access
Investigation and prediction of distributed volume rendering performance
(2017) Tkachev, Gleb
In this work, I describe the process of developing a cluster scalability model that is capable of predicting performance of a parallel rendering application running on a cluster while only having data that can be obtained from one of its nodes. I begin by studying scaling behavior of a single cluster, employing linear regression and neural networks to construct a cluster-specific scalability model, which im-plicitly captures its hardware characteristics. I use this model as a foundation for further work, developing a hardware-agnostic cluster scalability model. Instead of using explicit hardware characteristics as input, the hardware-agnostic model takes in a distribution of node computation time, which encapsulates local computational load of a rendering application, enabling the model to focus on pre-dicting communication overhead of a cluster. This allows simulation of different hardware by varying the node computation time, gathering enough data to train a neural network that predicts the overall performance of the rendering application on a cluster with arbitrary node hardware.
Open Access
Prediction and similarity models for visual analysis of spatiotemporal data
(2022) Tkachev, Gleb; Ertl, Thomas (Prof. Dr.)
Ever since the early days of computers, their usage have become essential in natural sciences. Whether through simulation, computer-aided observation or data processing, the progress in computer technology have been mirrored by the constant growth in the size of scientific data. Unfortunately, as the data sizes grow, and human capabilities remains constant, it becomes increasingly difficult to analyze and understand the data. Over the last decades, visualization experts have proposed many approaches to address the challenge, but even these methods have their limitations. Luckily, recent advances in the field of Machine Learning can provide the tools needed to overcome the obstacle. Machine learning models are a particularly good fit as they can both benefit from the large amount of data present in the scientific context and allow the visualization system to adapt to the problem at hand. This thesis presents research into how machine learning techniques can be adapted and extended to enable visualization of scientific data. It introduces a diverse set of techniques for analysis of spatiotemporal data, including detection of irregular behavior, self-supervised similarity metrics, automatic selection of visual representations and more. It also discusses the general challenges of applying Machine Learning to Scientific Visualization and how to address them.
Open Access
Visual analysis of droplet dynamics in large-scale multiphase spray simulations
(2021) Heinemann, Moritz; Frey, Steffen; Tkachev, Gleb; Straub, Alexander; Sadlo, Filip; Ertl, Thomas
We present a data-driven visual analysis approach for the in-depth exploration of large numbers of droplets. Understanding droplet dynamics in sprays is of interest across many scientific fields for both simulation scientists and engineers. In this paper, we analyze large-scale direct numerical simulation datasets of the two-phase flow of non-Newtonian jets. Our interactive visual analysis approach comprises various dedicated exploration modalities that are supplemented by directly linking to ParaView. This hybrid setup supports a detailed investigation of droplets, both in the spatial domain and in terms of physical quantities. Considering a large variety of extracted physical quantities for each droplet enables investigating different aspects of interest in our data. To get an overview of different types of characteristic behaviors, we cluster massive numbers of droplets to analyze different types of occurring behaviors via domain-specific pre-aggregation, as well as different methods and parameters. Extraordinary temporal patterns are of high interest, especially to investigate edge cases and detect potential simulation issues. For this, we use a neural network-based approach to predict the development of these physical quantities and identify irregularly advected droplets.