Next-generation HPC clusters are evolving into highly heterogeneous systems that integrate traditional computing resources with emerging accelerator technologies such as quantum processors, neuromorphic units, dataflow architectures, and specialized AI accelerators within a unified infrastructure. These advanced systems enable workloads to dynamically utilize different accelerators during various computation phases, creating complex execution patterns. The performance of the workloads can therefore be impacted by many factors, including how the accelerators are shared, their utilization, and their placement within the system. Moreover, effects such as the system and network state due to the overall system load can significantly impact the job completion rate. Understanding, identifying, and quantifying the impact of the most critical factors (e.g., the number of allocated accelerators) will help decide the investment decisions for accelerator acquisition and deployment that can improve the overall system throughput. This paper extensively studies these complex interactions among advanced accelerators within an HPC cluster and various workloads. We introduce a novel analytical model which predicts the speedup of a workload given an accelerator/system configuration. This model can be used to quantify the effect of augmenting additional accelerators on job performance running on an HPC cluster. We validate the model using both simulated and real environments.

In this work, we propose a framework to auto-tune the multiple network models' simulation configurations within SST/macro using Tree-structured Parzen Estimator-based Bayesian optimization to observe the effect on simulation accuracy across different message regimes. These regimes consist of small to large message sizes and latency to bandwidth-bound messages. We provide a detailed analysis of the simulation error for four representative HPC systems. Our Bayesian optimization-based autotuning framework for network models achieves a maximum of 5x improvement in accuracy over best-effort manual configurations based on available hardware specifications.

Hypergraphs are a popular paradigm to represent complex real-world networks exhibiting multi-way relationships of varying sizes. Mining centrality in hyper-graphs via symmetric adjacency tensors has only recently become computationally feasible for large and complex datasets. To enable scalable computation of these and related hypergraph analytics, here we focus on the Sparse Symmetric Tensor Times Same Vector (S3TTVC) operation. We introduce the Compound Compressed Sparse Symmetric (CCSS) format, an extension of the compact CSS format for hypergraphs of varying hyperedge sizes and present a shared-memory parallel algorithm to compute S3TTVC. We experimentally show S3TTVc computation using the CCSS format achieves better performance than the naive baseline, and is subsequently more performant for hypergraph$H$-eigenvector centrality.

In recent years, graph theoretic considerations have become increasingly important in the design of HPC interconnection topologies. One approach is to seek optimal or near-optimal families of graphs with respect to a particular graph theoretic property, such as diameter. In this work, we consider topologies which optimize the spectral gap. We study a novel HPC topology, SpectralFly, designed around the Ramanujan graph construction of Lubotzky, Phillips, and Sarnak (LPS). We show combinatorial properties, such as diameter, bisection bandwidth, average path length, and resilience to link failure, of SpectralFly topologies are better than, or comparable to, similarly constrained DragonFly, SlimFly, and BundleFly topologies. Additionally, we simulate the performance of SpectralFly on a representative sample of micro-benchmarks using the Structure Simulation Toolkit Macroscale Element Library simulator and study cost-minimizing layouts, demonstrating considerable benefit of the SpectralFly topology.

An airborne hyperspectral imager operating in the midwave-infrared spectral range is described. The Mid-infrared Airborne Hyperspectral Imager (MAHI) features 3.3-nm spectral sampling over its 3.3-5.4 μm wavelength range. MAHI operates in a roll-stabilized pushbroom configuration with 480 crosstrack pixels, each with an instantaneous field-of-view (IFOV) of 0.94 mrad, to provide for a total FOV of 25.8°. The sensor spectroradiometric performance is illustrated by case studies featuring the detection, identification, and quantification of a number of fugitive gaseous emissions from industrial sources.

A new airborne facility instrument for Earth science applications is introduced. The Mineral and Gas Identifier (MAGI) is a wide-swath (programmable up to ±42° off nadir) moderate spectral resolution thermal-infrared (TIR) imaging spectrometer that spans the 7.1- to 12.7-μm spectral window in 32 uniform and contiguous channels. Its spectral resolution enables improved discrimination of rock and mineral types, greatly expanded gas-detection capability, and generally more accurate land-surface temperature retrievals. The instrument design arose from trade studies between spectral resolution, spectral range, and instrument sensitivity and has now been validated by flight data acquired with the completed sensor. It offers a potential prototype for future space-based TIR instruments, which will require much higher spectral resolution than is currently available in order to address more detailed climate, anthropogenic, and solid Earth science questions.

Issues involved in operating a sophisticated scientific instrument as a computer peripheral accessible over a high-speed network are studied. A custom interactive visualization application was constructed to support investigation using a unique computer-controlled high-voltage electron microscope. The researcher's workstation forms the visible third of a triumvirate, along with the instrument and the compute resource. The software was designed to support not only image acquisition, but also many of the tasks that microscope researchers perform in analyzing images. The result of this case study is the identification of some of the issues regarding interacting with scientific instrumentation over high-speed networks and the construction of custom applications to support many of the tasks within a laboratory's research methodology.>