Snapshot Collection Research Articles

A globally convergent method to accelerate large-scale optimization using on-the-fly model hyperreduction: Application to shape optimization

We present a numerical method to efficiently solve optimization problems governed by large-scale nonlinear systems of equations, including discretized partial differential equations, using projection-based reduced-order models accelerated with hyperreduction (empirical quadrature) and embedded in a trust-region framework that guarantees global convergence. The proposed framework constructs a hyperreduced model on-the-fly during the solution of the optimization problem, which completely avoids an offline training phase. This ensures all snapshot information is collected along the optimization trajectory, which avoids wasting samples in remote regions of the parameters space that are never visited, and inherently avoids the curse of dimensionality of sampling in a high-dimensional parameter space. At each iteration of the proposed algorithm, a reduced basis and empirical quadrature weights are constructed precisely to ensure the global convergence criteria of the trust-region method are satisfied, ensuring global convergence to a local minimum of the original (unreduced) problem. Numerical experiments are performed on two fluid shape optimization problems to verify the global convergence of the method and demonstrate its computational efficiency; speedups over 18× (accounting for all computational cost, even cost that is traditionally considered “offline” such as snapshot collection and data compression) relative to standard optimization approaches that do not leverage model reduction are shown.

Jewish Women, English Dress: From Kindertransport to Women in Uniform in Wartime England

The study of dress in relation to Jewish identity in England is critically understudied. Historic anxieties concerning the presence of Jews in England impacted the lives of many, refugees seeking asylum from fascist Europe in particular. Images from the snapshot collection of Lorraine Sulzbacher, a German Jewish refugee, are correlated with oral history testimonies from other Kindertransport refugees to explore the identity formation and integration of female Kindertransport refugees in England from 1939 to 1945. This article investigates how dress was used as a visual signifier of integration, and how both dress and photographic practices simultaneously created space for identity formation. It discusses the impact dress had on the formation of identity during the transition from girlhood to adulthood for female Kindertransport refugees. Additionally, this article examines how uniformed war work provided these young women opportunities to develop their identities as displaced European Jewish persons in Britain. This analysis of dress and photography posits that these refugees were a diaspora in formation.

A digital-twin visualized architecture for Flexible Manufacturing System

The new generation of industrial 4.0 intelligent manufacturing system consists of Human-Cyber-Physical System (HCPS), integrating human with cyber and physical systems. In manufacturing, a digital-twin visualization architecture is to solve the human-machine interaction problem that concerns digital-twin modeling on the Cyber-Physical (C-P) side and on the Human-Cyber side. Although there are many related research and applications, there lacks attention in terms of full life cycle functional services and lightweight architecture. This paper presents a general architecture of digital-twin visualization for flexible manufacturing systems (FMS). How the digital-twin C-P modeling of multi-source heterogeneous information can be described is investigated and how the 3D visualized human-machine interaction with digital-twin scenario information is explored in the proposed architecture. Besides, the visualization method of high-value information, relating to the life cycle planning, design, debugging and service stages, is studied and discussed thoroughly. Also, a digital-twin modeling concept of "Geometric information (G)-Historical samples (H)-Object attribute (O)-Snapshot collection (S)-Topology constraint (T)" (GHOST) is proposed, and methods for developing virtual digital-twin scenes architecture are presented. Based on the proposed modeling concept of GHOST for digital-twin, prototypes have been developed for the general platform of digital-twin RESTful services and the cross-platform general visual mock-up software. Experimental results show that this method is effective in the FMS lifecycle in various aspects.

Evaluation of Galerkin and Petrov–Galerkin model reduction for finite element approximations of the shallow water equations

The shallow water equations (SWE) are widely used to model depth-averaged free-surface flows. They are well-studied, and many successful numerical methods have been posited for their solution. Nevertheless, accurate resolution of shallow-water flows can still be very computationally demanding. In particular, computational expense is a significant hurdle for inclusion of well-resolved shallow water models in engineering applications involving risk assessment, optimal design, or parameter estimation.The SWE are then a natural candidate for model reduction. In previous work by Lozovskiy et al. (2016), we presented an approach based on Proper Orthogonal Decomposition (POD) and hyper-reduction for a class of semi-implicit, stabilized finite element approximations of the SWE. Here, we revisit this scheme to further evaluate its consistency and performance using different snapshot collection procedures. We also consider a closely related Petrov–Galerkin technique based on least-squares minimization as an alternative to standard Galerkin projection. We formulate both schemes in a stabilized finite element context and examine their consistency rigorously both with and without hyper-reduction. We then evaluate their accuracy and robustness for test problems in one and two space dimensions.

Simulating Solute Transport in Porous Media Using Model Reduction Techniques

In this study, we introduce a numerical method to reduce the solute transport equation into a reduced form that can replicate the behavior of the model described by the original equation. The basic idea is to collect an ensemble of data of state variables (say, solute concentration), called snapshots, by running the original model, and then use the proper orthogonal decomposition (POD) techniques (or the Karhunen-Loeve decomposition) to create a set of basis functions that span the snapshot collection. The snapshots can be reconstructed using these basis functions. The solute concentration at any time and location in the domain is expressed as a linear combination of these basis functions, and a Galerkin procedure is applied to the original model to obtain a set of ordinary differential equations for the coefficients in the linear representation. The accuracy and computational efficiency of the reduced model have been demonstrated using several one-dimensional and two-dimensional examples

Real Time Snapshot Collection Algorithm for Mobile Distributed Systems with Minimum Number of Checkpoints

Checkpointing is an efficient way of implementing fault tolerance in distributed systems. Mobile computing raises many new issues, such as high mobility, lack of stable storage on mobile hosts (MHs), low bandwidth of wireless channels, limited battery life and disconnections that make the traditional checkpointing protocols unsuitable for such systems. Minimum process non-blocking coordinated checkpointing may be useful for mobile distributed system as this approach is domino-free, requires at most two checkpoints of each process on stable storage, forces only interacting processes to checkpoint and does not suspend their underlying computation during checkpointing. Sometimes, it also requires piggybacking of information onto normal messages, blocking of the underlying computation or taking some useless checkpoints. In this paper, we propose a non-blocking minimum process coordinated checkpointing algorithm that requires minimum bandwidth over wireless channels and does not requires any induced/forced or mutable checkpoints and reduce the height of checkpointing tree without taking any extra overhead in real time.

Modeling and control of physical processes using proper orthogonal decomposition

The proper orthogonal decomposition (POD) technique (or the Karhunan Loève procedure) has been used to obtain low-dimensional dynamical models of many applications in engineering and science. In principle, the idea is to start with an ensemble of data, called snapshots, collected from an experiment or a numerical procedure of a physical system. The POD technique is then used to produce a set of basis functions which spans the snapshot collection. When these basis functions are used in a Galerkin procedure, they yield a finite-dimensional dynamical system with the smallest possible degrees of freedom. In this context, it is assumed that the physical system has a mathematical model, which may not be available for many physical and/or industrial applications. In this paper, we consider the steady-state Rayleigh-Bénard convection whose mathematical model is assumed to be unknown, but numerical data are available. The aim of the paper is to show that, using the obtained ensemble of data, POD can be used to model accurately the natural convection. Furthermore, this approach is very efficient in the sense that it uses the smallest possible number of parameters, and thus, is suited for process control. Particularly, we consider two boundary control problems 1. (a) tracking problem, and 2. (b) avoiding hot spot in a certain region of the domain.

Results and challenges in Web search evaluation

A frozen 18.5 million page snapshot of part of the Web has been created to enable and encourage meaningful and reproducible evaluation of Web search systems and techniques. This collection is being used in an evaluation framework within the Text Retrieval Conference (TREC) and will hopefully provide convincing answers to questions such as, “Can link information result in better rankings?”, “Do longer queries result in better answers?”, and, “Do TREC systems work well on Web data?” The snapshot and associated evaluation methods are described and an invitation is extended to participate. Preliminary results are presented for an effectivess comparison of six TREC systems working on the snapshot collection against five well-known Web search systems working over the current Web. These suggest that the standard of document rankings produced by public Web search engines is by no means state-of-the-art.

City, State, and Public Ritual in the Late-Medieval Burgundian Netherlands

At the end of a distinguished career as chronicler of the Burgundian court, Georges Chastellain (1404–75) penned a quick sketch of the outstanding accomplishments of his duke, Charles the Bold. Accustomed to expositions awash in chivalric pomp, Chastellain employed a different tack to commemorate this sovereign: He sketched eleven “magnificences” performed by the duke of Burgundy, all reconstructed images of this prince's engagement with ceremony. Foremost among this snapshot collection of state ritual was neither a tournament, nor a wedding ceremony, nor even a processional entry. What stood out, in Chastellain's estimation, as Charles' greatest deed was something more riveting and more powerful than any of these spectacles so beloved by the fifteenth-century Burgundian court:The first [magnificence] was at Brussels, where, seated on his throne, his sword unsheathed and held by his Marshall, he gathered the men of Ghent arranged kneeling before him and at his pleasure and in their presence cut and tore up the political charters they bore. Done for permanent record, this action was without parallel.For Chastellain, the supreme magnificence of Charles the Bold was a lesson in exemplary punishment, the public abasement of the aldermen and guild deans of the Flemish city of Ghent in January 1469, a year and a half after a city revolt of rank-and-file guildsmen had unsettled celebrations in honor of his accession to the countship of Flanders.

Low-cost checkpointing and failure recovery in mobile computing systems

A mobile computing system consists of mobile and stationary nodes, connected to each other by a communication network. The presence of mobile nodes in the system places constraints on the permissible energy consumption and available communication bandwidth. To minimize the lost computation during recovery from node failures, periodic collection of a consistent snapshot of the system (checkpoint) is required. Locating mobile nodes contributes to the checkpointing and recovery costs. Synchronous snapshot collection algorithms, designed for static networks, either force every node in the system to take a new local snapshot, or block the underlying computation during snapshot collection. Hence, they are not suitable for mobile computing systems. If nodes take their local checkpoints independently in an uncoordinated manner, each node may have to store multiple local checkpoints in stable storage. This is not suitable for mobile nodes as they have small memory. This paper presents a synchronous snapshot collection algorithm for mobile systems that neither forces every node to take a local snapshot, nor blocks the underlying computation during snapshot collection. If a node initiates snapshot collection, local snapshots of only those nodes that have directly or transitively affected the initiator since their last snapshots need to be taken. We prove that the global snapshot collection terminates within a finite time of its invocation and the collected global snapshot is consistent. We also propose a minimal rollback/recovery algorithm in which the computation at a node is rolled back only if it depends on operations that have been undone due to the failure of node(s). Both the algorithms have low communication and storage overheads and meet the low energy consumption and low bandwidth constraints of mobile computing systems.