As computational capabilities increasingly outpace disk speeds on leading supercomputers, scientists will, in turn, be increasingly unable to save their simulation data at its native resolution. One solution to this problem is to compress these data sets as they are generated and visualize the compressed results afterwards. We explore this approach, specifically subsampling velocity data and the resulting errors for particle advection-based flow visualization. We compare three techniques: random selection of subsamples, selection at regular locations corresponding to multi-resolution reduction, and introduce a novel technique for informed selection of subsamples. Furthermore, we explore an adaptive system which exchanges the subsampling budget over parallel tasks, to ensure that subsampling occurs at the highest rate in the areas that need it most. We perform supercomputing runs to measure the effectiveness of the selection and adaptation techniques. Overall, we find that adaptation is very effective, and, among selection techniques, our informed selection provides the most accurate results, followed by the multi-resolution selection, and with the worst accuracy coming from random subsamples.

This work focuses on enabling the use of Python-based methods for the purpose of performing in situ analysis and visualization. This approach facilitates access to and use of a rapidly growing collection of Python-based, third-party libraries for analysis and visualization, as well as lowering the barrier to entry for user-written Python analysis codes. Beginning with a simulation code that is instrumented to use the SENSEI in situ interface, we present how to couple it with a Python-based data consumer, which may be run in situ, and in parallel at the same concurrency as the simulation. We present two examples that demonstrate the new capability. One is an analysis of the reaction rate in a proxy simulation of a chemical reaction on a 2D substrate, while the other is a coupling of an AMR simulation to Yt, a parallel visualization and analysis library written in Python. In the examples, both the simulation and Python in situ method run in parallel on a large-scale HPC platform.

Measurements of absolute runtime are useful as a summary of performance when studying parallel visualization and analysis methods on computational platforms of increasing concurrency and complexity. We can obtain even more insights by measuring and examining more detailed measures from hardware performance counters, such as the number of instructions executed by an algorithm implemented in a particular way, the amount of data moved to/from memory, memory hierarchy utilization levels via cache hit/miss ratios, and so forth. This work focuses on performance analysis on modern multi-core platforms of three different visualization and analysis kernels that are implemented in different ways: one is "traditional", using combinations of C++ and VTK, and the other uses a data-parallel approach using VTK-m. Our performance study consists of measurement and reporting of several different hardware performance counters on two different multi-core CPU platforms. The results reveal interesting performance differences between these two different approaches for implementing these kernels, results that would not be apparent using runtime as the only metric.

Fluid mechanics considers two frames of reference for an observerwatching a flow field: Eulerian and Lagrangian. The former isthe frame of reference traditionally used for flow analysis, and involvesextracting particle trajectories based on a vector field. Withthis work, we explore the opportunities that arise when consideringthese trajectories from the Lagrangian frame of reference. Specifically,we consider a form where flows are extracted in situ and thenused for subsequent post hoc analysis. We believe this alternate,Lagrangian-based form will be increasingly useful, because the Eulerianframe of reference is sensitive to temporal frequency, andarchitectural trends are causing temporal frequency to drop rapidlyon modern supercomputers. We support our viewpoint by running aseries of experiments, which demonstrate the Lagrangian form canbe more accurate, require less I/O, and be faster when compared totraditional advection.

This work examines performance characteristics of multiple shared-memory implementations of a probabilistic graphical modeling (PGM) optimization code, which forms the basis for an advanced, state-of-the art image segmentation method. The work is motivated by the need to accelerate scientific image analysis pipelines in use by experimental science, such as at x-ray light sources, and is motivated by the need for platform-portable codes that perform well across many different computational architectures. The primary focus of this work and its main contribution is an in-depth study of shared-memory parallel performance of different implementations, which include those using alternative parallelization approaches such as C11-threads, OpenMP, and data parallel primitives (DPPs). Our results show that, for this complex data-intensive algorithm, the DPP implementation exhibits better runtime performance, but also exhibits less favorable scaling characteristics than the C11-threads and OpenMP counterparts. Based upon a set of experiments that collect hardware performance counters on multiple platforms, the reason for the runtime performance difference appears to be due primarily to algorithmic efficiency gains: the reformulation from the traditional C11-threads and OpenMP expression of the solution into that of data parallel primitives results in significantly fewer instructions being executed. This study is the first of its type to do performance analysis using hardware counters for comparing methods based on VTK-m-based data-parallel primitives with those based on more traditional OpenMP or threads-based parallelism. It is timely, as there is increasing awareness of the need for platform portability in light of increasing node-level parallelism and increasing device heterogeneity.