Control and Data Plane Security of High Performance Networks
Explore the source record for details and available documents.
97 MATHEMATICS AND COMPUTING↗
SEARCH · Search NASA
Results for “High Performance Networking”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Understanding and Avoiding Performance Variability in High Performance Networks.
Abstract not provided.
Grant, Ryan↗
Interpretable Models for Workflow Differentiation in High-Performance Scientific Networks
Scientific workflows in high-performance networks spawn hundreds of interdependent flows that must be managed collectively—yet existing network classifiers treat each flow in isolation, leading to fragmented QoS decisions and missed interflow patterns. We present a novel traffic classification solution that operates at the workflow level, distinguishing entire filetransfer operations from streaming analytics by capturing how concurrent flows interact and burst together. We introduce a workflow identification window (WIW) that ingests raw packet headers from parallel flows into unified tensors, preserving the spatial-temporal patterns that differentiate scientific workflows. This approach achieves 98.7% accuracy using CNN, LSTM, and hybrid architectures, while maintaining 84% accuracy on production traffic collected a week later—demonstrating robustness to temporal drift. By integrating SHAP and GradCAM explainability, we reveal that early-packet timing patterns and cross-flow correlations drive classification decisions, providing operators with interpretable insights. Our system enables coherent workflow-level QoS enforcement and dynamic bandwidth allocation in scientific networks, eliminating manual per-flow configuration while maintaining classification latency at millisecond level.
Giannakou, Anna [LBL, Berkeley]↗
Computing Bottleneck Structures at Scale for High-Precision Network Performance Analysis
The Theory of Bottleneck Structures is a recently-developed framework for studying the performance of data networks. It describes how local perturbations in one part of the network propagate and interact with others. This framework is a powerful analytical tool that allows network operators to make accurate predictions about network behavior and thereby optimize performance. Previous work implemented a software package for bottleneck structure analysis, but applied it only to toy examples. In this work, we introduce the first software package capable of scaling bottleneck structure analysis to production-size networks. Here, we benchmark our system using logs from ESnet, the Department of Energy's high-performance data network that connects research institutions in the U.S. Using the previously published tool as a baseline, we demonstrate that our system achieves vastly improved performance, constructing the bottleneck structure graphs in 0.21 s and calculating link derivatives in 0.09 s on average. We also study the asymptotic complexity of our core algorithms, demonstrating good scaling properties and strong agreement with theoretical bounds. These results indicate that our new software package can maintain its fast performance when applied to even larger networks. They also show that our software is efficient enough to analyze rapidly changing networks in real time. Overall, we demonstrate the feasibility of applying bottleneck structure analysis to solve practical problems in large, real-world data networks.
benchmark↗
SCinet Architecture: Featured at the International Conference for High Performance Computing,Networking, Storage and Analysis 2016
SCinet is the purpose-built network that operates during the International Conference for High Performance Computing,Networking, Storage and Analysis (Super Computing or SC). Created each year for the conference, SCinet brings to life a high-capacity network that supports applications and experiments that are a hallmark of the SC conference. The network links the convention center to research and commercial networks around the world. This resource serves as a platform for exhibitors to demonstrate the advanced computing resources of their home institutions and elsewhere by supporting a wide variety of applications. Volunteers from academia, government and industry work together to design and deliver the SCinet infrastructure. Industry vendors and carriers donate millions of dollars in equipment and services needed to build and support the local and wide area networks. Planning begins more than a year in advance of each SC conference and culminates in a high intensity installation in the days leading up to the conference. The SCinet architecture for SC16 illustrates a dramatic increase in participation from the vendor community, particularly those that focus on network equipment. Software-Defined Networking (SDN) and Data Center Networking (DCN) are present in nearly all aspects of the design.
Lyonnais, Marc↗
Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training
In this paper, we extend an existing runtime system (the TensorFlow runtime) to enable automatic concurrency control and scheduling of operations. We explore performance modeling to predict the performance of operations with various thread- level parallelism. Our performance model is highly accurate and lightweight. Leveraging the performance model, our runtime system employs a set of scheduling strategies that co-run opera- tions to improve hardware utilization and system throughput. Our runtime system demonstrates a significant performance benefit. Comparing with using the recommended configurations for concurrency control and operation scheduling in TensorFlow, our approach achieves 36% performance (execution time) im- provement on average (up to 49%) for four neural network models, and achieves high performance close to the optimal one manually obtained by the user.
Liu, Jiawen↗
A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers
Current Deep Learning models use highly optimized convolutional neural networks (CNN) trained on large graphical processing units (GPU)-based computers with a fairly simple layered network topology, i.e., highly connected layers, without intra-layer connections. Complex topologies have been proposed, but are intractable to train on current systems. Building the topologies of the deep learning network requires hand tuning, and implementing the network in hardware is expensive in both cost and power.In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing (HPC) to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. Due to input size limitations of current quantum computers we use the MNIST dataset for our evaluation.The results show the possibility of using the three architectures in tandem to explore complex deep learning networks that are untrainable using a von Neumann architecture. We show that a quantum computer can find high quality values of intra-layer connections and weights, while yielding a tractable time result as the complexity of the network increases; a high performance computer can find optimal layer-based topologies; and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware.This represents a new capability that is not feasible with current von Neumann architecture. It potentially enables the ability to solve very complicated problems unsolvable with current computing technologies.
97 MATHEMATICS AND COMPUTING↗
A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers
Current Deep Learning models use highly optimized convolutional neural networks (CNN) trained on large graphical processing units (GPU)-based computers with a fairly simple layered network topology, i.e., highly connected layers, without intra-layer connections. Complex topologies have been proposed, but are intractable to train on current systems. Building the topologies of the deep learning network requires hand tuning, and implementing the network in hardware is expensive in both cost and power. In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing (HPC) to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. Due to input size limitations of current quantum computers we use the MNIST dataset for our evaluation. The results show the possibility of using the three architectures in tandem to explore complex deep learning networks that are untrainable using a von Neumann architecture. We show that a quantum computer can find high quality values of intra-layer connections and weights, while yielding a tractable time result as the complexity of the network increases; a high performance computer can find optimal layer-based topologies; and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware. This represents a new capability that is not feasible with current von Neumann architecture. It potentially enables the ability to solve very complicated problems unsolvable with current computing technologies.
Potok, Thomas E↗
Flexspander: augmenting expander networks in high-performance systems with optical bandwidth steering
Communication efficiency is one of the deciding factors in determining many of today’s high-performance computing (HPC) applications. Traditionally, HPC systems have been on static network topologies, making them inflexible to the variety of skewed traffic demands that may arise due to the spatial locality inherent in many applications. To handle traffic locality, researchers have proposed integrating optical circuit switches (OCSs) into the network architecture, which reconfigures the network topology to alter and dynamically adapt to the predicted traffic. In this paper, we present a novel reconfigurable network topology called Flexspander. Beyond offering a flexible interconnect, Flexspander also offers full flexibility in terms of construction and can be built with any arbitrary combination of commercial electrical packet switches and OCSs. We evaluate Flexspander performance through extensive simulations with multiple network traces, and our results show improved performance for Flexspander over currently proposed static and reconfigurable topologies in terms of the flow completion time.
Teh, Min Yee↗
A three-dimensional carbon nano-network for high performance lithium ion batteries
Three-dimensional (3D) network structure has been envisioned as a superior architecture for lithium ion battery (LIB) electrodes, which enhances both ion and electron transport to significantly improve battery performance. Herein, a 3D carbon nano-network is fabricated through chemical vapor deposition of carbon on a scalably manufactured 3D porous anodic alumina (PAA) template. As a demonstration on the applicability of 3D carbon nano-network for LIB electrodes, the low conductivity active material, TiO 2 , is then uniformly coated on the 3D carbon nano-network using atomic layer deposition. High power performance is demonstrated in the 3D C/TiO 2 electrodes, where the parallel tubes and gaps in the 3D carbon nano-network facilitates fast Li ion transport. A large areal capacity of ~0.37 mAh·cm –2 is achieved due to the large TiO 2 mass loading in the 60 µm-thick 3D C/TiO 2 electrodes. At a test rate of C/5, the 3D C/TiO 2 electrode with 18 nm-thick TiO 2 delivers a high gravimetric capacity of ~240 mAh g –1 , calculated with the mass of the whole electrode. A long cycle life of over 1000 cycles with a capacity retention of 91% is demonstrated at 1C. In this study, the effects of the electrical conductivity of carbon nano-network, ion diffusion, and the electrolyte permeability on the rate performance of these 3D C/TiO 2 electrodes are systematically studied.
25 ENERGY STORAGE↗
Reliability Lessons Learned From GPU Experience With The Titan Supercomputer at Oak Ridge Leadership Computing Facility
The high computational capability of graphics processing units (GPUs) is enabling and driving the scientific discovery process at large-scale. The world’s second fastest supercomputer for open science, Titan, has more than 18,000 GPUs that computational scientists use to perform scientific simu- lations and data analysis. Understanding of GPU reliability characteristics, however, is still in its nascent stage since GPUs have only recently been deployed at large-scale. This paper presents a detailed study of GPU errors and their impact on system operations and applications, describing experiences with the 18,688 GPUs on the Titan supercom- puter as well as lessons learned in the process of efficient operation of GPUs at scale. These experiences are helpful to HPC sites which already have large-scale GPU clusters or plan to deploy GPUs in the future.
Tiwari, Devesh↗
Multi-Tenant Isolation via Reconfigurable Networks
High performance computing environments are often used for a wide variety of workloads ranging from simulation, data transformation and analysis, and complex workflows to name just a few. These systems may process data at various security levels but in so doing are often enclaved at the highest security posture. This approach places significant restrictions on the users of the system even when processing data at a lower security level and exposes data at higher levels of confidentiality to a much broader population than otherwise necessary. The traditional approach of isolation, while effective in establishing security enclaves poses significant challenges for the use of shared infrastructure in HPC environments. This report details current state-of-the-art in reconfigurable network enclaving through Software Defined Networking (SDN) and Network Function Virtualization (NFV) and their applicability to secure enclaves in HPC environments. SDN and NFV methods are based on a solid foundation of system wide virtualization. The purpose of which is very straight forward, the system administrator can deploy networks that are more amenable to customer needs, and at the same time achieve increased scalability making it easier to increase overall capacity as needed without negatively affecting functionality. The network administration of both the server system and the virtual sub-systems is simplified allowing control of the infrastructure through well-defined APIs (Application Programming Interface). While SDN and NFV technologies offer significant promise in meeting these goals, they also provide the ability to address a significant component of the multi-tenant challenge in HPC environments, namely resource isolation. Traditional HPC systems are built upon scalable high-performance networking technologies designed to meet specific application requirements. Dynamic isolation of resources within these environments has remained difficult to achieve. SDN and NFV methodology provide us with relevant concepts and available open standards based APIs that isolate compute and storage resources within an otherwise common networking infrastructure. Additionally, the integration of the networking APIs within larger system frameworks such as OpenStack provide the tools necessary to establish isolated enclaves dynamically allowing the benefits of HPC while providing a controlled security structure surrounding these systems.
97 MATHEMATICS AND COMPUTING↗
A multidisciplinary approach to the development of low-cost high-performance lightwave networks
Our research focuses on high-speed distributed systems. We anticipate that our results will allow the fabrication of low-cost networks employing multi-gigabit-per-second data links for space and military applications. The recent development of high-speed low-cost photonic components and new generations of microprocessors creates an opportunity to develop advanced large-scale distributed information systems. These systems currently involve hundreds of thousands of nodes and are made up of components and communications links that may fail during operation. In order to realize these systems, research is needed into technologies that foster adaptability and scaleability. Self-organizing mechanisms are needed to integrate a working fabric of large-scale distributed systems. The challenge is to fuse theory, technology, and development methodologies to construct a cost-effective, efficient, large-scale system.
Maitan, Jacek↗
High-Performance, Semi-Interpenetrating Polymer Network
High-performance polymer made by new synthesis in which one or more easy-to-process, but brittle, thermosetting polyimides combined with one or more tough, but difficult-to-process, linear thermoplastics to yield semi-interpenetrating polymer network (semi-IPN) having combination of easy processability and high tolerance to damage. Two commercially available resins combined to form tough, semi-IPN called "LaRC-RP49." Displays improvements in toughness and resistance to microcracking. LaRC-RP49 has potential as high-temperature matrix resin, adhesive, and molding resin. Useful in aerospace, automotive, and electronic industries.
Pater, Ruth H.↗
High Speed Vertical Cavity Surface Emitting Lasers for High Performance Optical Networks
Explore the source record for details and available documents.
97 MATHEMATICS AND COMPUTING↗
Nuclear Physics Network Requirements Review Report
The Energy Sciences Network (ESnet) is the Office of Science’s high-performance network user facility, delivering highly reliable data transport capabilities optimized for the requirements of data-intensive science. In essence, ESnet is the circulatory system that enables the U.S. Department of Energy (DOE) science mission by connecting each and every DOE lab and its user facilities. ESnet is funded and stewarded by the Advanced Scientific Computing Research (ASCR) Program and managed and operated by the Scientific Networking Division at Lawrence Berkeley National Laboratory (LBNL). ESnet is widely regarded as a global leader in the research and education networking community. ESnet connects DOE national laboratories, user facilities, and major experiments so scientists can use remote instruments and computing resources as well as share data with collaborators, transfer large data sets, and access distributed data repositories. While ESnet provides network connectivity, it cannot be characterized as an internet service provider as it is specifically built to provide a range of network services that are tailored to meet the unique requirements of DOE’s data-intensive science.
97 MATHEMATICS AND COMPUTING↗