Scalable Software Framework Research Articles

HIPPYlib-MUQ: A Bayesian Inference Software Framework for Integration of Data with Complex Predictive Models under Uncertainty

Bayesian inference provides a systematic framework for integration of data with mathematical models to quantify the uncertainty in the solution of the inverse problem. However, the solution of Bayesian inverse problems governed by complex forward models described bypartial differential equations (PDEs)remains prohibitive with black-boxMarkov chain Monte Carlo (MCMC)methods. We present hIPPYlib-MUQ, an extensible and scalable software framework that contains implementations of state-of-the art algorithms aimed to overcome the challenges of high-dimensional, PDE-constrained Bayesian inverse problems. These algorithms accelerate MCMC sampling by exploiting the geometry and intrinsic low-dimensionality of parameter space via derivative information and low rank approximation. The software integrates two complementary open-source software packages, hIPPYlib and MUQ. hIPPYlib solves PDE-constrained inverse problems using automatically-generated adjoint-based derivatives, but it lacks full Bayesian capabilities. MUQ provides a spectrum of powerful Bayesian inversion models and algorithms, but expects forward models to come equipped with gradients and Hessians to permit large-scale solution. By combining these two complementary libraries, we created a robust, scalable, and efficient software framework that realizes the benefits of each and allows us to tackle complex large-scale Bayesian inverse problems across a broad spectrum of scientific and engineering disciplines. To illustrate the capabilities of hIPPYlib-MUQ, we present a comparison of a number of MCMC methods available in the integrated software on several high-dimensional Bayesian inverse problems. These include problems characterized by both linear and nonlinear PDEs, various noise models, and different parameter dimensions. The results demonstrate that large (∼ 50×) speedups over conventional black box and gradient-based MCMC algorithms can be obtained by exploiting Hessian information (from the log-posterior), underscoring the power of the integrated hIPPYlib-MUQ framework.

Korali: Efficient and scalable software framework for Bayesian uncertainty quantification and stochastic optimization

We present Korali, an open-source framework for large-scale Bayesian uncertainty quantification and stochastic optimization. The framework relies on non-intrusive sampling of complex multiphysics models and enables their exploitation for optimization and decision-making. In addition, its distributed sampling engine makes efficient use of massively-parallel architectures while introducing novel fault tolerance and load balancing mechanisms. We demonstrate these features by interfacing Korali with existing high-performance software such as Aphros, LAMMPS (CPU-based), and Mirheo (GPU-based) and show efficient scaling for up to 512 nodes of the CSCS Piz Daint supercomputer. Finally, we present benchmarks demonstrating that Korali outperforms related state-of-the-art software frameworks.

Parallel general purpose multiobjective optimization framework with application to electron beam dynamics

Particle accelerators are invaluable tools for research in the basic and applied sciences, such as materials science, chemistry, the biosciences, particle physics, nuclear physics and medicine. The design, commissioning, and operation of accelerator facilities is a nontrivial task, due to the large number of control parameters and the complex interplay of several conflicting design goals. The Argonne Wakefield Accelerator facility has some unique challenges resulting from its purpose to carry out advanced accelerator R Individual experiments often have challenging beam requirements, and the physical configuration of the beam lines is often changed to accommodate the variety of supported experiments. The need for rapid deployment of different operational settings further complicates the optimization work that must be done for multiple constraints and challenging operational regimes. One example of this is an independently staged two-beam acceleration experiment which requires the construction of an additional beam line (this is now in progress). The high charge drive beam, well into the space charge regime, must be threaded through small aperture (17.6 mm) decelerating structures. In addition, the bunch length must be sufficiently short to maximize power generation in the decelerator. We propose to tackle this problem by means of multiobjective optimization algorithms which also facilitate a parallel deployment. In order to compute solutions in a meaningful time frame, a fast and scalable software framework is required. In this paper, we present a general-purpose framework for simulation-based multiobjective optimization methods that allows the automatic investigation of optimal sets of machine parameters. Using evolutionary algorithms as the optimizer and opal as the forward solver, validation experiments and results of multiobjective optimization problems in the domain of beam dynamics are presented. Optimized solutions for the new high charge drive beam line found by the framework were used to finish the design of a two beam acceleration experiment. The selected solution along with the associated beam parameters is presented.

A scalable software framework for data integration in bioprocess development.

Effectiveness in lab workflows-despite progresses made in automation and lab informatics-is often hindered by insufficient integration of devices and data. The iLAB software framework, a middleware connecting and integrating devices and data, provides a plugin architecture that can be adapted to individual lab environments. Integration of devices is preferably based on standardized data and communication protocols. In addition to device integration, process data and result data from different sources (e.g. readers) are converted to a standard format and administered by a powerful database for further processing. In this paper, the use of iLAB in a bioprocess development application is described. Process parameters and measured values of two high-throughput bioreactor systems (96-well plates and 10 mL reactor vessels) are collected in the iLAB database. Data from screening experiments and offline data are visualized, analyzed, and compared. A filter algorithm allows searching for matching parameters in experiments as well as the comparison of correlated datasets, independently from the used bioreactor system. The database model enables the consolidation of data, the transition from data to information, and a solid base for management decisions on enterprise level.

Control Software Framework ECSF for Experiment Based on CORBA and Agent

The control system for Inertia Confinement Fusion (ICF) experiment facility is based on a scalable software framework and can implement centralized control of thousands of devices and roboticized experiment processes. Experiment Control Software Framework (ECSF) is the control software framework specially for large-scale experiment facility, colligating advantages of CORBA and Agent. By the reuse at the system level, control software can be fleetly constructed. Software device Agent based on OOD is the base of framework. Centralized manager and common module implement events, alerts, reservations, logging, data archive, flow and arithmetic management. Several connection managements are also provided for the member of the framework. Taking patch energy subsystem as an example, the instantiation of the framework is introduced to validate the feasibility and validity of ECSF.

A software framework for data dimensionality reduction: application to chemical crystallography

AbstractMaterials science research has witnessed an increasing use of data mining techniques in establishing process‐structure‐property relationships. Significant advances in high‐throughput experiments and computational capability have resulted in the generation of huge amounts of data. Various statistical methods are currently employed to reduce the noise, redundancy, and the dimensionality of the data to make analysis more tractable. Popular methods for reduction (like principal component analysis) assume a linear relationship between the input and output variables. Recent developments in non‐linear reduction (neural networks, self‐organizing maps), though successful, have computational issues associated with convergence and scalability. Another significant barrier to use dimensionality reduction techniques in materials science is the lack of ease of use owing to their complex mathematical formulations. This paper reviews various spectral‐based techniques that efficiently unravel linear and non‐linear structures in the data which can subsequently be used to tractably investigate process‐structure‐property relationships. In addition, we describe techniques (based on graph‐theoretic analysis) to estimate the optimal dimensionality of the low‐dimensional parametric representation. We show how these techniques can be packaged into a modular, computationally scalable software framework with a graphical user interface ‐ Scalable Extensible Toolkit for Dimensionality Reduction (SETDiR). This interface helps to separate out the mathematics and computational aspects from the materials science applications, thus significantly enhancing utility to the materials science community. The applicability of this framework in constructing reduced order models of complicated materials dataset is illustrated with an example dataset of apatites described in structural descriptor space. Cluster analysis of the low‐dimensional plots yielded interesting insights into the correlation between several structural descriptors like ionic radius and covalence with characteristic properties like apatite stability. This information is crucial as it can promote the use of apatite materials as a potential host system for immobilizing toxic elements.

A software framework for data dimensionality reduction: application to chemical crystallography

Materials science research has witnessed an increasing use of data mining techniques in establishing process‐structure‐property relationships. Significant advances in high‐throughput experiments and computational capability have resulted in the generation of huge amounts of data. Various statistical methods are currently employed to reduce the noise, redundancy, and the dimensionality of the data to make analysis more tractable. Popular methods for reduction (like principal component analysis) assume a linear relationship between the input and output variables. Recent developments in non‐linear reduction (neural networks, self‐organizing maps), though successful, have computational issues associated with convergence and scalability. Another significant barrier to use dimensionality reduction techniques in materials science is the lack of ease of use owing to their complex mathematical formulations. This paper reviews various spectral‐based techniques that efficiently unravel linear and non‐linear structures in the data which can subsequently be used to tractably investigate process‐structure‐property relationships. In addition, we describe techniques (based on graph‐theoretic analysis) to estimate the optimal dimensionality of the low‐dimensional parametric representation. We show how these techniques can be packaged into a modular, computationally scalable software framework with a graphical user interface ‐ Scalable Extensible Toolkit for Dimensionality Reduction (SETDiR). This interface helps to separate out the mathematics and computational aspects from the materials science applications, thus significantly enhancing utility to the materials science community. The applicability of this framework in constructing reduced order models of complicated materials dataset is illustrated with an example dataset of apatites described in structural descriptor space. Cluster analysis of the low‐dimensional plots yielded interesting insights into the correlation between several structural descriptors like ionic radius and covalence with characteristic properties like apatite stability. This information is crucial as it can promote the use of apatite materials as a potential host system for immobilizing toxic elements.

A cloud computing based framework for general 2D and 3D cellular automata simulation

Cellular automata can be applied to solve several problems in a variety of areas, such as biology, chemistry, medicine, physics, astronomy, economics, and urban planning. The automata are defined by simple rules that give rise to behavior of great complexity running on very large matrices. 2D applications may require more than 10^6x10^6 matrix cells, which are usually beyond the computational capacity of local clusters of computers. This paper presents a solution for traditional cellular automata simulations. We propose a scalable software framework, based on cloud computing technology, which is capable of dealing with very large matrices. The use of the framework facilitates the instrumentation of simulation experiments by non-computer experts, as it removes the burden related to the configuration of MapReduce jobs, so that researchers need only be concerned with their simulation algorithms.

A fast and scalable low dimensional solver for charged particle dynamics in large particle accelerators

Particle accelerators are invaluable tools for research in the basic and applied sciences, in fields such as materials science, chemistry, the biosciences, particle physics, nuclear physics and medicine. The design, commissioning, and operation of accelerator facilities is a non-trivial task, due to the large number of control parameters and the complex interplay of several conflicting design goals.We propose to tackle this problem by means of multi-objective optimization algorithms which also facilitate massively parallel deployment. In order to compute solutions in a meaningful time frame, that can even admit online optimization, we require a fast and scalable software framework. In this paper, we focus on the key and most heavily used component of the optimization framework, the forward solver. We demonstrate that our parallel methods achieve a strong and weak scalability improvement of at least two orders of magnitude in today’s actual particle beam configurations, reducing total time to solution by a substantial factor.Our target platform is the Blue Gene/P (Blue Gene/P is a trademark of the International Business Machines Corporation in the United States, other countries, or both) supercomputer. The space-charge model used in the forward solver relies significantly on collective communication. Thus, the dedicated TREE network of the platform serves as an ideal vehicle for our purposes. We demonstrate excellent strong and weak scalability of our software which allows us to perform thousands of forward solves in a matter of minutes, thus already allowing close to online optimization capability.

Software mechanisms for extensible and scalable 3D visualization of construction operations

This paper presents research that led to the design and implementation of an extensible and scalable software framework for the dynamic 3D visualization of simulated construction operations. In the domain of operations design and analysis, the ability to see a 3D animation of processes that have been simulated allows for two very important things: verification and validation. In addition, a model can be communicated effectively which, coupled with verification and validation, makes it “credible” and thus used in making decisions. In the presented research, a set of core animation statements of the most general use from the extension viewpoint is first identified. Second, methods to design an add-on interface to the identified core animation methods are investigated by capitalizing on documented principles of application framework design. Finally, the designed add-on interface and its scalability are validated by implementing the extensible framework on multiple computing platforms and then extending the language with several non-trivial extensions using the designed add-on interface. The research concluded that geometric transformation-based animation statements are: (1) collectively sufficient to visually describe a broad class of common construction processes, and (2) are at a level of abstraction that they can be logically concatenated to describe higher-level motion dynamics involved in performing construction. In addition, it was found that an open, loosely-coupled visualization scheme and direct interface methods to append the animation statement interpreter’s vocabulary with new add-on designed statements: (1) allows language extensions without modification to or understanding of the underlying methods, and (2) presents users with a consistent interface to visually describe construction processes, thereby providing complexity concealment and a uniform end-user interface. The presented framework is implemented as an extension (add-on) interface to the VITASCOPE visualization system.

Intelligent resource discovery using ontology-based resource profiles

Successful resource discovery across heterogeneous repositories is highly dependent on the semantic and syntactic homogeneity of the associated resource descriptions in each repository. Ideally, consistent resource descriptions are easily extracted from each repository, expressed using standard syntactic and semantic structures, and managed and accessed within a distributed, flexible, and scalable software framework. In practice however, seldom do all three of these elements exist. To help address this situation, the Object Oriented Data Technology (OODT) project at the Jet Propulsion Laboratory has developed an extensible, standards-based resource description scheme that provides the necessary description and management facilities for the discovery of resources across heterogeneous repositories. The OODT resource description scheme can be used across scientific domains to describe any resource. It uses a small set of generally accepted, broadly-scoped descriptors while also providing a mechanism for the inclusion of domain-specific descriptors. In addition, the OODT scheme can be used to capture hierarchical, relational and recursive relationships between resources. In this paper we expand on prior work and describe an intelligent resource discovery framework that consists of separate software and data architectures focusing on the standard resource description scheme. We illustrate intelligent resource discovery using a case study that provides efficient search across distributed repositories using common interfaces and a hierarchy of resource descriptions derived from a complex, domain-specific ontology.