Search NASA⌕ Search

SEARCH · Search NASA

Results for “Packet losses”

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.

100 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

LEED: A Lightwave Energy-Efficient Datacenter

The Lightwave Energy-Efficient Datacenter (LEED) program is a disruptive “green-field” approach that provides a quantum leap in the energy efficiency of datacenters. LEED’s fundamental value proposition is that a novel and re-architected optical network—RotorNet— can deliver “more bandwidth per buck” as well as unique system-level attributes that significantly improve overall datacenter energy efficiency and performance. LEED has developed three system-level testbeds. The first testbed uses calibrated hardware and software power measurements to determine server energy efficiency as a function of network bandwidth and workload. These measurements have shown that increasing network communications bandwidth dramatically increases server energy efficiency providing a realistic path to the overall ENLITENED program goal of doubling the number of transactions per joule. The second testbed demonstrates key hardware: a prototype low-loss, high-port count optical “selector switch”. This switch was fabricated, racked, and tested. Measured switch characteristics include loss, bandwidth, crosstalk, switch time, system-level switch time (including the transceivers), and bit error rate. The third testbed demonstrates a fully working and manufactured pinwheel design which dramatically lowers the cost of design, while delivering high switch radix and low reconfiguration times. The LEED project has tied these three novel photonic switch prototypes together with production servers and software through the development of a novel FPGA-based NIC platform called Corundum. Corundum ensures that the packet-switched protocols supported by commodity operating systems and devices can interface with the Rotor switch design. The LEED group has used this combined hardware and software prototype to characterize applications running at a commercially relevant scale. The project has used a combination of enhanced optical modulation amplitude (OMA) modulators, broadband multiplexers and demultiplexers, avalanche photodiodes, and a novel burst-mode receivers to enable the insertion of LEED-developed optical switches without the need for expensive optical amplification. Our modeling has shown that measured LEED-developed device characteristics can achieve link characteristics of 2 pJ/bit including both transceivers and the Rotor switch. In summary, the LEED program has demonstrated a credible and practical path, through novel hardware and software, to realize the program objectives of ENLITENED. The net result will ensure that the United States maintains its strength in the crucial sector of Information Technology, which is vital to both our economic security and our national security.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Real-Time Inference For MI/RR Deblending

The Fermilab Main Injector (MI) and Recycler Ring (RR) share a common beam loss monitor (BLM) system, making loss events difficult to attribute to their source machine when beam is present in both simultaneously. The Real-time Edge AI for Distributed Systems (READS) project addresses this by deblending BLM readings in real time using machine learning (ML). The current FPGA based implementation meets the sub-3 ms latency requirement but carries a resource intensive hls4ml development cycle, motivating exploration of GPU based deployment. This paper characterizes inference latency on an NVIDIA Jetson Orin Nano and introduces a packet organization scheme for assembling synchronized event frames from seven distributed BLM DAQ streams. Using a Python based DAQ simulation with injected timing jitter in place of unavailable live beam data, the pipeline achieved an average end to end latency of 0.456 ms (σ = 0.122 ms) across 167,000 test frames, comfortably meeting the timing constraint. Early outliers were attributed to TensorRT warm-up rather than steady state limitations, suggesting GPU based inference is a viable alternative to the existing FPGA implementation.

Yu, Kellen [Cornell U.]↗
DOE OSTI2026

Exploration of Real Time Inference for MI-RR Deblending on GPU/TPU Systems

The Fermilab Main Injector (MI) and Recycler Ring (RR) share a common beam loss monitor (BLM) system, making loss events difficult to attribute to their source machine when beam is present in both simultaneously. The Real-time Edge AI for Distributed Systems (READS) project addresses this by deblending BLM readings in real time using machine learning (ML). The current FPGA based implementation meets the sub-3 ms latency requirement but carries a resource intensive hls4ml development cycle, motivating exploration of GPU based deployment. This paper characterizes inference latency on an NVIDIA Jetson Orin Nano and introduces a packet organization scheme for assembling synchronized event frames from seven distributed BLM DAQ streams. Using a Python based DAQ simulation with injected timing jitter in place of unavailable live beam data, the pipeline achieved an average end to end latency of 0.456 ms (σ = 0.122 ms) across 167,000 test frames, comfortably meeting the timing constraint. Early outliers were attributed to TensorRT warm-up rather than steady state limitations, suggesting GPU based inference is a viable alternative to the existing FPGA implementation.

Yu, Kellen [Fermilab; Cornell U.]↗
NASA NTRS2002

Overview of the High Performance Antiproton (HiPAT) Experiment

The annihilation of matter with antimatter represents the highest energy density of any known reaction, producing 10(exp 8) MJ/g, approximately 10 orders of magnitude more energy per unit mass than chemical based combustion. To take the first step towards using this energy for propulsion applications the NASA MSFC Propulsion Research Center (PRC) has initiated a research activity examining the storage of low energy antiprotons. Storage was identified as a key enabling technology since it builds the experience base necessary to understand the handling of antiprotons for virtually all utilization and high-density storage concepts. To address this need, a device referred to as the High Performance Antiproton Trap (HiPAT) is under development at the NASA MSFC PRC. The HiPAT is an electromagnetic system (Penning-Malmberg design) consisting of a 4 Tesla superconductor, a high voltage confinement electrode system (operation up to 20 KV), and an ultra high vacuum test section (operating in the 10(exp -12) torr range). The system was designed to be portable with an ultimate goal of maintaining 10(exp 12) charged particles with a half-life of 18 days. Currently, this system is being experimentally evaluated using normal matter ions which are cheap to produce and relatively easy to handle. These normal ions provide a good indication of overall trap behavior, with the exception of assessing annihilation losses. The ions are produced external to HiPAT using two hydrogen ion sources, with adjustable beam energy and current. Ion are transported in a beam line and controlled through the use of electrostatic optics. These optics serve to both focus and gate the incoming ions, providing microsecond-timed pulses that are dynamically captured by cycling the HiPAT electric containment field like a 'trap door'. The layout of this system more closely simulates the operations expected at an actual antiproton production facility where 'packets' of antiprotons with pulse widths measured in 100's of nanoseconds could be provided. Initial dynamic capture experiments have been performed with both trap and ton source system functioning at approximately 10% of maximum levels, minimizing the potential for extraneous effects. Dynamic trapping techniques demonstrated the successful capture of millions of hydrogen ions with good agreement with the predicted loading based on the timing sequence, trap electric field, and ion beam current. These techniques will be expanded to examine multiple shot capture or 'stacking' to increase the total number of ions stored within HiPAT.

Martin, James J.↗
NASA NTRS2003

Improving TCP Network Performance by Detecting and Reacting to Packet Reordering

There are many factors governing the performance of TCP-basec applications traversing satellite channels. The end-to-end performance of TCP is known to be degraded by the reordering, delay, noise and asymmetry inherent in geosynchronous systems. This result has been largely based on experiments that evaluate the performance of TCP in single flow tests. While single flow tests are useful for deriving information on the theoretical behavior of TCP and allow for easy diagnosis of problems they do not represent a broad range of realistic situations and therefore cannot be used to authoritatively comment on performance issues. The experiments discussed in this report test TCP s performance in a more dynamic environment with competing traffic flows from hundreds of TCP connections running simultaneously across the satellite channel. Another aspect we investigate is TCP's reaction to bit errors on satellite channels. TCP interprets loss as a sign of network congestion. This causes TCP to reduce its transmission rate leading to reduced performance when loss is due to corruption. We allowed the bit error rate on our satellite channel to vary widely and tested the performance of TCP as a function of these bit error rates. Our results show that the average performance of TCP on satellite channels is good even under conditions of loss as high as bit error rates of 10(exp -5)

Kruse, Hans↗
NASA NTRS1976

Two-photon coherent states of the radiation field

The concept of a two-photon coherent state is introduced for applications in quantum optics. It is a simple generalization of the well-known minimum-uncertainty wave packets. The detailed properties of two-photon coherent states are developed and distinguished from ordinary coherent states. These two-photon coherent states are mathematically generated from coherent states through unitary operators associated with quadratic Hamiltonians. Physically they are the radiation states of ideal two-photon lasers operating far above threshold, according to the self-consistent-field approximation. The mean-square quantum noise behavior of these states, which is basically the same as those of minimum-uncertainty states, leads to applications not obtainable from coherent states or one-photon lasers. The essential behavior of two-photon coherent states is unchanged by small losses in the system. The counting rates or distributions these states generate in photocount experiments also reveal their difference from coherent states.

Yuen, H. P.↗
DOE OSTI2024

Collapse of neutrino wave functions under Penrose gravitational reduction

Models of spontaneous wave function collapse have been postulated to address the measurement problem in quantum mechanics. Their primary function is to convert coherent quantum superpositions into incoherent ones, with the result that macroscopic objects cannot be placed into widely separated superpositions for observably prolonged times. Many of these processes will also lead to loss of coherence in neutrino oscillations, producing observable signatures in the flavor profile of neutrinos at long travel distances. The majority of studies of neutrino oscillation coherence to date have focused on variants of the continuous state localization model, whereby an effective decoherence strength parameter is used to model the rate of coherence loss with an assumed energy dependence. Another class of collapse models that have been proposed posit connections to the configuration of gravitational field accompanying the mass distribution associated with each wave function that is in the superposition. A particularly interesting and prescriptive model is Penrose’s description of gravitational collapse which proposes a decoherence time τ determined through E g τ ∼ ℏ , where E g is a calculable function of the Newtonian gravitational potential. Here we explore application of the Penrose collapse model to neutrino oscillations, reinterpreting previous experimental limits on neutrino decoherence in terms of this model. We identify effects associated with both spatial collapse and momentum diffusion, finding that the latter is ruled out in data from the IceCube South Pole Neutrino Observatory so long as the neutrino wave packet width at production is σ ν , x ≤ 2 × 10 − 12 m . Published by the American Physical Society 2024

Astronomy & Astrophysics↗
DOE OSTI2023

Evaluating Ion Accumulation and Storage in Traveling Wave Based Structures for Lossless Ion Manipulations

Structures for lossless ion manipulations (SLIM) technology has demonstrated high resolving power ion mobility separation and flexibility to integrate complex ion manipulations into a single experimental platform. To enable IMS separations, trapping/accumulating ions inside SLIM (or in-SLIM) prior to injection of a packet for separations provides ease of operation and reduces the need for dedicated ion traps external to SLIM. To fully characterize the ion accumulation process, we have evaluated the effect of TW amplitudes, ion collection times, and storage times on the “in-SLIM” accumulation process. The study utilized a SLIM module comprising 5 distinct tracks, each with a specific ion accumulation configuration. The effect of the TW conditions on the accumulation process was investigated for a 3-peptide mixture: kemptide, angiotensin II, and neurotensin at a TW speed of 106 m/s. The effect of ion accumulation time/collection time and storage time was investigated, in addition to TW amplitude. Overall, the signal of the analyte ions increased when the ion collection time increased from 49 to 163 ms but decreased when the ion collection time increased further to 652 ms due to the space charge effects. Ion losses were observed at high TW amplitudes (e.g., 15 V p–p and 20 V p–p ). In addition, under space charge conditions (e.g., collection times of 163 and 652 ms), the signal of the analyte ions decreased with an increase in storage times for all TW amplitudes applied to the trapping region. For ion accumulation, the data indicate that gentler TW conditions must be utilized to minimize ion losses and fragments to benefit from the “in-SLIM” accumulation process. Finally, wider SLIM tracks provided better performance than those with narrower tracks.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Computational Offload with BlueField Smart NICs

The recent introduction of a new generation of "smart NICs" have provided new accelerator platforms that include CPU cores or reconfigurable fabric in addition to traditional networking hardware and packet offloading capabilities. While there are currently several proposals for using these smartNICs for low-latency, in-line packet processing operations, there remains a gap in knowledge as to how they might be used as computational accelerators for traditional high-performance applications. This work aims to look at benchmarks and mini-applications to evaluate possible benefits of using a smartNIC as a compute accelerator for HPC applications. We investigate NVIDIA's current-generation BlueField-2 card, which includes eight Arm CPUs along with a small amount of storage, and we test the networking and data movement performance of these cards compared to a standard Intel server host. We then detail how two different applications, YASK and miniMD can be modified to make more efficient use of the BlueField-2 device with a focus on overlapping computation and communication for operations like neighbor building and halo exchanges. Our results show that while the overall compute performance of these devices is limited, using them with a modified miniMD algorithm allows for potential speedups of 5 to 20% over the host CPU baseline with no loss in simulation accuracy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Sentinel

Network intrusion detection systems (NIDS) are commonplace in network security but they frequently employ algorithms that are computational demanding requiring hardware and software with significant power requirements. Two examples of such resource-intensive algorithms used for network security are regular expression matching and broader signature pattern matching which are commonly used in deep packet inspection (DPI). Network security algorithms that have large power requirements may be a challenge for low-power internet-of-things (IoT) environments, which generally lack the power resources to implement complex security measures like computationally expensive DPI at the edge. Furthermore, IoT environments incorporating 5G standalone networks have network latency constraints beyond just power that make DPI at the edge even more difficult. Programmable logic is ideally suited for machine learning inference for DPI because of its deep instruction level parallelism and single-cycle memory access. Machine learning approaches for DPI have been explored before using the programmable logic of field programmable gate arrays (FPGA) as a potential solution for NIDS approaches that would be power-suitable for IoT. However, those previous programmable logic NIDS approaches utilize either a supervised or unsupervised learning model. Sentinel utilizes the ensemble of these two machine learning approaches known as a semi-supervised approach which has shown promise in NIDS implementations. Sentinel provides a programmable logic implementation of a semi-supervised approach for DPI which operates at much lower power and latency than a GPU implementation with negligible loss of accuracy due to quantization through a logistic regressor.

Anderson, MatthewW [Idaho National Laboratory (INL↗