Search NASA⌕ Search

SEARCH · Search NASA

Results for “COMMUNICATIONS NETWORKS”

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.

At least 1,189 records · Page 66

External beam planning module of Eclipse for external beam radiation therapy

Eclipse is a 3-dimensional (3D) treatment planning system for radiation therapy offered by Varian Medical Systems, Inc. The system has the network connectivity for the electronic transfer of image datasets and digital data communication among different equipment. The scope of this project for this special issue of Medical Dosimetry on 3D treatment planning systems is the assessment of planning tools in the external beam planning module of Eclipse to generate optimized treatment plans for patients undergoing external beam radiation therapy. This treatment planning system is relatively mature to be able to generate (1) simple treatment plans, (2) conformal radiation therapy plans, (3) static intensity-modulated radiation therapy (IMRT) plans, (4) volumetric-modulated arc therapy (VMAT) plans, and (5) treatment plans for electron beam therapy. The treatment planning tools are relatively plentiful to assist in the radiation therapy treatment planning. Some new features have been incorporated in the latest version and are helpful for making high-quality treatment plans. However, the location of the tools is not intuitive, and hence, familiarity with the user interface is essential to the efficient use of the treatment planning system. In addition, there are a number of dose algorithms available for the computation of dose distributions. The understanding of each dose computation algorithm is essential for the optimal use of this treatment planning system.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genome Atlas (TCGA) database. WSI images can be in the size of 100,000 x 100,000 pixels or even larger, and in this work we studied the effect of image resolution on a classification task, while achieving state-of-the-art AUC scores. Moreover, our approach doesn't need pixel-level labels, since we're avoiding patching from the WSI images completely, while adding the capability of training arbitrary large-size images. This is achieved through a distributed spatial decomposition method, by leveraging the non-block fat-tree interconnect network of the Summit architecture, which enabled GPU-to-GPU direct communication. Finally, detailed performance analysis results are shown, as well as a comparison with a data-parallel approach when possible.

Tsaris, Aristeidis (aris)↗

A Latency Tolerant Runtime System for Mass Market Computer Systems

Grappa is a latency tolerant runtime system for mass market computer systems comprised of multiple nodes having x86 multicore processors and infiniband interconnect. Grappa is designed to enable in-memory irregular applications written in the C++ programming language to scale up on multiprocessor systems. Grappa addresses the load balance, communication, and synchronization challenges posed by running applications that do not partition evenly across large systems. Two mechanisms are essential to its success: fast task switching and message aggregation. Fast task switching enables processors to tolerate communication and synchronization delays; aggregation sustains high bandwidth on low injection rate networks.

Source record↗

Evaluating Trade-offs in Potential Exascale Interconnect Topologies

This study looks at the high-level trade-offs for interconnect architectures that are likely to be available in HPC machines in the 2021-2022 timeframe. The study uses multiple simulation models to look at the effects of varying machine size, injection bandwidth and topology for a selected set of benchmark communication patterns. The studies show the sensitivity of various applications to these network parameters. This study can help inform decisions on node size versus number of nodes, which topology is most efficient for our applications, how much injection bandwidth to provision, etc. These simulations only look at interconnects with maximally configured global bandwidth, and studies of networks configured with less will be the subject of future milestones.

97 MATHEMATICS AND COMPUTING↗

Modular Security Apparatus for Managing Distributed Cryptography for Command-and-Control Messages on Operational Technology Networks (Module-OT)

Module-OT is a bump- in- the- wire solution acting as a secure conduit for data between devices or systems across a network. Using the latest in open-source cryptographic libraries, all defined communications undergo authentication, authorization, and encryption. The core system can be easily installed through industry standard processes, and the use of popular open-source packages allows for it to be customizable customized by the developer community or deployed in unique embedded environments.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Bringing Control System User Interfaces to the Web

With the evolution of web based technologies, especially HTML5 [1], it becomes possible to create web-based control system user interfaces (UI) that are cross-browser and cross-device compatible. This article describes two technologies that facilitate this goal. The first one is the WebOPI [2], which can seamlessly display CSS BOY [3] Operator Interfaces (OPI) in web browsers without modification to the original OPI file. The WebOPI leverages the powerful graphical editing capabilities of BOY and provides the convenience of re-using existing OPI files. On the other hand, it uses generic JavaScript and a generic communication mechanism between the web browser and web server. It is not optimized for a control system, which results in unnecessary network traffic and resource usage. Our second technology is the WebSocket-based Process Data Access (WebPDA) [4]. It is a protocol that provides efficient control system data communication using WebSocket [5], so that users can create web-based control system UIs using standard web page technologies such as HTML, CSS and JavaScript. WebPDA is control system independent, potentially supporting any type of control system.

Chen, Xihui↗

A Latency Tolerant Runtime System for Mass Market Computer Systems

Grappa is a latency tolerant runtime system for mass market computer systems comprised of multiple nodes having x86 multicore processors and infiniband interconnect. Grappa is designed to enable in-memory irregular applications written in the C++ programming language to scale up on multiprocessor systems. Grappa addresses the load balance, communication, and synchronization challenges posed by running applications that do not partition evenly across large systems. Two mechanisms are essential to its success: fast task switching and message aggregation. Fast task switching enables processors to tolerate communication and synchronization delays; aggregation sustains high bandwidth on low injection rate networks.

Source record↗

Distributed gather/scatter operations across a network of memory nodes

Devices, methods, and systems for distributed gather and scatter operations in a network of memory nodes. A responding memory node includes a memory; a communications interface having circuitry configured to communicate with at least one other memory node; and a controller. The controller includes circuitry configured to receive a request message from a requesting node via the communications interface. The request message indicates a gather or scatter operation, and instructs the responding node to retrieve data elements from a source memory data structure and store the data elements to a destination memory data structure. The controller further includes circuitry configured to transmit a response message to the requesting node via the communications interface. The response message indicates that the data elements have been stored into the destination memory data structure.

97 MATHEMATICS AND COMPUTING↗

DragonView: Toward Understanding Network Interference in Dragonfly-based Supercomputers

Acquiring and maintaining high performance computing systems represents a substantial, long term investment. Therefore, it is paramount to optimize their utilization in order to maximize the return on investment and the impact on science and engineering. In this context, network congestion and inter-job interference are a major challenge as they can lead to significant systemwide performance degregation. This is especially pertinent in shared interconnects such as the dragonfly network, in which resources are shared by multiple jobs. Since many large-scale applications are communication-bound, a few communication heavy applications can have a large impact on the performance of all jobs and decrease the overall system throughput. In this paper, we describe a web-based visual analytics environment to study congestion and inter-job interference on dragonfly networks. The system enables network architects, system administrators and application developers to explore the potential impact of machine configuration, job placement policies and routing algorithms by visualizing and comparing ensembles of simulated workloads and real world data. We focus on the design of the system and describe the challenges, design decisions and the final implementation. Finally, we present two detailed case studies of prototypical analysis sessions by our stakeholders aimed at exploring the implications of different workloads and machine configurations using ensembles of network simulations.

97 MATHEMATICS AND COMPUTING↗

Modeling Large-Scale Slim Fly Networks Using Parallel Discrete-Event Simulation

As supercomputers approach exascale performance, the increased number of processors translates to an increased demand on the underlying network interconnect. We present that the slim fly network topology, a new low-diameter, low-latency, and low-cost interconnection network, is gaining interest as one possible solution for next-generation supercomputing interconnect systems. In this article, we present a high-fidelity slim fly packet-level model leveraging the Rensselaer Optimistic Simulation System (ROSS) and Co-Design of Exascale Storage (CODES) frameworks. We validate the model with published work before scaling the network size up to an unprecedented 1 million compute nodes and confirming that the slim fly observes peak network throughput at extreme scale. In addition to synthetic workloads, we evaluate large-scale slim fly models with real communication workloads from applications in the Design Forward program with over 110,000 MPI processes. We show strong scaling of the slim fly model on an Intel cluster achieving a peak network packet transfer rate of 2.3 million packets per second and processing over 7 billion discrete events using 128 MPI tasks. Enabled by the strong performance capabilities of the model, we perform a detailed application trace and routing protocol performance study. Lastly, through analysis of metrics such as packet latency, hop count, and congestion, we find that the slim fly network is able to leverage simple minimal routing and achieve the same performance as more complex adaptive routing for tested DOE benchmark applications.

97 MATHEMATICS AND COMPUTING↗

Porting GASNet to Portals: Partitioned Global Address Space (PGAS) Language Support for the Cray XT

Partitioned Global Address Space (PGAS) Languages are an emerging alternative to MPI for HPC applications development. The GASNet library from Lawrence Berkeley National Lab and the University of California at Berkeley provides the network runtime for multiple implementations of four PGAS Languages: Unified Parallel C (UPC), Co-Array Fortran (CAF), Titanium and Chapel. GASNet provides a low overhead one-sided communication layer has enabled portability and high performance of PGAS languages. This paper describes our experiences porting GASNet to the Portals network API on the Cray XT series.

97 MATHEMATICS AND COMPUTING↗

Control coordination in inverter‐based microgrids using AoI‐based 5G schedulers

Abstract A coordinated set point automatic adjustment with correction enabled (C‐SPAACE) framework that uses 5G communication for real‐time control coordination between inverter‐based resources (IBR) in microgrids is proposed. Utilising slicing capability, 5G offers low‐latency communication to C‐SPAACE under normal conditions. However, given the multitude of power grid use cases, a certain 5G slice for C‐SPAACE may have access only to limited radio spectrum resources, which if not managed well, greatly undermines the communication needs of C‐SPAACE framework. Thus, optimally scheduling the available spectrum resources among IBRs in a sliced 5G network‐based C‐SPAACE framework becomes a critical problem. To address this issue, the authors utilise a novel age of information (AoI) metric and designs an AoI‐based 5G scheduler to provide low‐latency communication to C‐SPAACE. Following this, a co‐simulation environment is designed using PSCAD/EMTDC and Python to simulate a microgrid supported by 5G communication. Time‐domain simulation case studies are performed using the proposed co‐simulation environment to evaluate the performance of C‐SPAACE using 5G with both AoI‐based and other baseline (non‐AoI) schedulers.

Choudhury, Biplav↗

Optomechanical transducers for quantum-information processing

We discuss the implementation of optical quantum networks where the interface between stationary and photonic qubits is realized by optomechanical transducers [K. Stannigel et al., Phys. Rev. Lett. 105, 220501 (2010)]. This approach does not rely on the optical properties of the qubit and thereby enables optical quantum communication applications for a wide range of solid-state spin- and charge-based systems. We present an effective description of such networks for many qubits and give a derivation of a state transfer protocol for long-distance quantum communication. We also describe how to mediate local on-chip interactions by means of the optomechanical transducers that can be used for entangling gates. We finally discuss experimental systems for the realization of our proposal.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Slides for LDRD review presentation of 18A12-131FP, "Cyber-Secure Physical Layer Security in Millimeter Wave Communications"

Objective Enhance cyber security, protection, and assurance of advanced mmWave wireless communications in the presence of unauthorized devices and intruders with novel capabilities within the physical layer. Significance Wireless communications using millimeter waves (mmWave) in the GigaHertz (GHz) band is a key new technology that will play a major role in the 5th Generation (5G) wireless networks. Cyber-physical solutions for other frequency bands are not directly applicable to mmWave technologies. We propose to validate the hypotheses that: 1) communications can be better protected from eavesdropping and cyber intrusion with security features built within the mmWave physical layer, 2) User authentication can be conducted in a non-intrusive manner and with high probability of success by the use of physical channel variations of mmWaves which get scattered by a human body.

99 GENERAL AND MISCELLANEOUS↗

HPC Interconnection Networks: The Key to Exascale Computing

Scientists from many domains desire to address problems within the next decade that, by all estimates, require computer systems that can achieve sustained exaflop computing rates (i.e., 1 x 10{sup 18} floating point operations per second) with real-world applications. Simply scaling existing designs is insufficient: analysis of current technological trends suggests that only a few architectural components are on track to reach the performance levels needed for exascale computing. The network connecting computer system nodes presents a particularly difficult challenge because of the prevalence of a wide variety of communication patterns and collective communication operations in algorithms used in scientific applications and their tendency to be the most significant limit to application scalability. Researchers at Oak Ridge National Laboratory and elsewhere are actively working to overcome these network-related scalability barriers using advanced hardware and software design, alternative network topologies, and performance prediction using modeling and simulation.

97 MATHEMATICS AND COMPUTING↗

Topolgy Agnostic Hot-Spot Avoidance with InfiniBand

InfiniBand has become a very popular interconnect, due to its advanced features and open standard. Large scale InfiniBand clusters are becoming very popular, as reflected by the TOP 500 supercomputer rankings. However, even with popular topologies like constant bi-section bandwidth Fat Tree, hot-spots may occur with InfiniBand, due to inappropriate configuration of network paths, presence of other jobs in the network and un-availability of adaptive routing. In this paper, we present a hot-spot avoidance layer (HSAL) for InfiniBand, which provides hot-spot avoidance using path bandwidth estimation and multi-pathing using LMC mechanism, without taking the network topology into account. We propose an adaptive striping policy with batch based striping and sorting approach, for efficient utilization of disjoint network paths. Integration of HSAL with MPI, the de facto programming model of clusters, shows promising results with collective communication primitives and MPI applications.

97 MATHEMATICS AND COMPUTING↗

Exploiting communication concurrency on high performance computing systems

Although logically available, applications may not exploit enough instantaneous communication concurrency to maximize hardware utilization on HPC systems. This is exacerbated in hybrid programming models such as SPMD+OpenMP. We present the design of a "multi-threaded" runtime able to transparently increase the instantaneous network concurrency and to provide near saturation bandwidth, independent of the application configuration and dynamic behavior. The runtime forwards communication requests from application level tasks to multiple communication servers. Our techniques alleviate the need for spatial and temporal application level message concurrency optimizations. Experimental results show improved message throughput and bandwidth by as much as 150% for 4KB bytes messages on InfiniBand and by as much as 120% for 4KB byte messages on Cray Aries. For more complex operations such as all-to-all collectives, we observe as much as 30% speedup. This translates into 23% speedup on 12,288 cores for a NAS FT implemented using FFTW. We also observe as much as 76% speedup on 1,500 cores for an already optimized UPC+OpenMP geometric multigrid application using hybrid parallelism.

97 MATHEMATICS AND COMPUTING↗