Search NASA⌕ Search

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.

At least 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

OpenSHMEM-UCX : Evaluation of UCX for implementing OpenSHMEM Programming Model

The OpenSHMEM reference implementation was developed towards the goal of developing an open source and high-performing Open- SHMEM implementation. To achieve portability and performance across various networks, the OpenSHMEM reference implementation uses GAS- Net and UCCS for network operations. Recently, new network layers have emerged with the promise of providing high-performance, scalabil- ity, and portability for HPC applications. In this paper, we implement the OpenSHMEM reference implementation to use the UCX framework for network operations. Then, we evaluate its performance and scalabil- ity on Cray XK systems to understand UCX s suitability for developing the OpenSHMEM programming model. Further, we develop a bench- mark called SHOMS for evaluating the OpenSHMEM implementation. Our experimental results show that OpenSHMEM-UCX outperforms the vendor supplied OpenSHMEM implementation in most cases on the Cray XK system by up to 40% with respect to message rate and up to 70% for the execution of application kernels.

Baker, Matthew B↗
DOE OSTI2019

Evaluating Quality of Service Traffic Classes on the Megafly Network

An emerging trend in High Performance Computing (HPC) systems that use hierarchical topologies (such as dragonfly) is that the applications are increasingly exhibiting high run-to-run performance variability. This poses a significant challenge for application developers, job schedulers, and system maintainers. One approach to address the performance variability is to use newly proposed network topologies such as megafly (or dragonfly+) that offer increased path diversity compared to a traditional fully connected dragonfly. Yet another approach is to use quality of service (QoS) traffic classes that ensure bandwidth guarantees. In this work, we select HPC application workloads that have exhibited performance variability on current 2-D dragonfly systems. We evaluate the baseline performance expectations of these workloads on megafly and 1-D dragonfly network models with comparably similar network configurations. Our results show that the megafly network, despite using fewer virtual channels (VCs) for deadlock avoidance than a dragonfly, performs as well as a fully connected 1-D dragonfly network. We then exploit the fact that megafly networks require fewer VCs to incorporate QoS traffic classes. We use bandwidth capping and traffic differentiation techniques to introduce multiple traffic classes in megafly networks. In some cases, our results show that QoS can completely mitigate application performance variability while causing minimal slowdown to the background network traffic.

Mubarak, Misbah↗
NASA NTRS2017

Classification of Wildfires from MODIS Data Using Neural Networks

Wildfires are destructive to both life and property, which necessitates an approach to quickly and autonomously detect these events from orbital observatories. This talk will introduce a neural network based approach for classifying wildfires in MODIS multispectral data, and will show how it could be applied to a constellation of low-cost CubeSats. The approach combines training a deep neural network on the ground using high performance consumer GPUs, with a highly optimized inference system running on a flight-proven embedded processor. Normally neural networks execute on hardware orders of magnitude more powerful than anything found in a space-based computer, therefore the inference system is designed to be performance even on the most modest of platforms. This implementation is able to be significantly more accurate than previous neural network implementations, while also approaching the accuracy of the state-of-the-art MODFIRE data products.

Artificial Intelligence↗
DOE OSTI2024

CommBench: Micro-Benchmarking Hierarchical Networks with Multi-GPU, Multi-NIC Nodes

Modern high-performance computing systems have multiple GPUs and network interface cards (NICs) per node. The resulting network architectures have multilevel hierarchies of subnetworks with different interconnect and software technologies. These systems offer multiple vendor-provided communication capabilities and library implementations (IPC, MPI, NCCL, RCCL, OneCCL) with APIs providing varying levels of performance across the different levels. Understanding this performance is currently difficult because of the wide range of architectures and programming models (CUDA, HIP, OneAPI). We present CommBench, a library with cross-system portability and a high-level API that enables developers to easily build microbenchmarks relevant to their use cases and gain insight into the performance (bandwidth & latency) of multiple implementation libraries on different networks. We demonstrate CommBench with three sets of microbenchmarks that profile the performance of six systems. Our experimental results reveal the effect of multiple NICs on optimizing the bandwidth across nodes and also present the performance characteristics of four available communication libraries within and across nodes of NVIDIA, AMD, and Intel GPU networks.

Hidayetoglu, Mert↗
DOE OSTI2020

Toward lower-diameter large-scale HPC and data center networks with co-packaged optics

We investigate the advantages of using co-packaged optics for building low-diameter, large-scale high-performance computing (HPC) and data center networks. The increased escape bandwidth offered by co-packaged optics can enable high-radix switch implementations of more than 150 switch ports, which can be combined with data rates of up to 400 Gb/s per port. From the network architecture perspective, the key benefits of using co-packaged optics in future fat-tree networks include (a) the ability to implement large-scale topologies of > <#comment/> 11 , 000 end points by eliminating the need for a third switching layer and (b) the ability to provide up to 4 × <#comment/> higher bisection bandwidth compared to existing solutions, reducing at the same time the number of required switch application-specific integrated circuits by > <#comment/> 80 % <#comment/> . From the network operation perspective, both reduced energy consumption and lower packet delays can be achieved since fewer hops are required; i.e., packets need to traverse fewer serializer/deserializer lanes and fewer switch buffers, which reduces the probability of contending with other packets and improves the tolerance of network congestion. The performance of the proposed architecture is evaluated via discrete-event simulations for a wide range of representative HPC synthetic-traffic cases that include both hotspot and non-hotspot scenarios. The simulation results suggest that co-packaged optics form a promising solution to keep up with bandwidth scaling in future networks, while the reduced number of switching layers can lead to significant mean packet delay improvements that start from 30% and reach up to 74% for high-load conditions.

Maniotis, Pavlos (ORCID:0000000244905253)↗
DOE OSTI2023

Catalyst: Fast and flexible modeling of reaction networks

We introduce Catalyst.jl, a flexible and feature-filled Julia library for modeling and high-performance simulation of chemical reaction networks (CRNs). Catalyst supports simulating stochastic chemical kinetics (jump process), chemical Langevin equation (stochastic differential equation), and reaction rate equation (ordinary differential equation) representations for CRNs. Through comprehensive benchmarks, we demonstrate that Catalyst simulation runtimes are often one to two orders of magnitude faster than other popular tools. More broadly, Catalyst acts as both a domain-specific language and an intermediate representation for symbolically encoding CRN models as Julia-native objects. This enables a pipeline of symbolically specifying, analyzing, and modifying CRNs; converting Catalyst models to symbolic representations of concrete mathematical models; and generating compiled code for numerical solvers. Leveraging ModelingToolkit.jl and Symbolics.jl, Catalyst models can be analyzed, simplified, and compiled into optimized representations for use in numerical solvers. Finally, we demonstrate Catalyst’s broad extensibility and composability by highlighting how it can compose with a variety of Julia libraries, and how existing open-source biological modeling projects have extended its intermediate representation.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2015

TTEthernet for Integrated Spacecraft Networks

Aerospace projects have traditionally employed federated avionics architectures, in which each computer system is designed to perform one specific function (e.g. navigation). There are obvious downsides to this approach, including excessive weight (from so much computing hardware), and inefficient processor utilization (since modern processors are capable of performing multiple tasks). There has therefore been a push for integrated modular avionics (IMA), in which common computing platforms can be leveraged for different purposes. This consolidation of multiple vehicle functions to shared computing platforms can significantly reduce spacecraft cost, weight, and design complexity. However, the application of IMA principles introduces significant challenges, as the data network must accommodate traffic of mixed criticality and performance levels - potentially all related to the same shared computer hardware. Because individual network technologies are rarely so competent, the development of truly integrated network architectures often proves unreasonable. Several different types of networks are utilized - each suited to support a specific vehicle function. Critical functions are typically driven by precise timing loops, requiring networks with strict guarantees regarding message latency (i.e. determinism) and fault-tolerance. Alternatively, non-critical systems generally employ data networks prioritizing flexibility and high performance over reliable operation. Switched Ethernet has seen widespread success filling this role in terrestrial applications. Its high speed, flexibility, and the availability of inexpensive commercial off-the-shelf (COTS) components make it desirable for inclusion in spacecraft platforms. Basic Ethernet configurations have been incorporated into several preexisting aerospace projects, including both the Space Shuttle and International Space Station (ISS). However, classical switched Ethernet cannot provide the high level of network determinism required by real-time spacecraft applications. Even with modern advancements, the uncoordinated (i.e. event-driven) nature of Ethernet communication unavoidably leads to message contention within network switches. The arbitration process used to resolve such conflicts introduces variation in the time it takes for messages to be forwarded. TTEthernet1 introduces decentralized clock synchronization to switched Ethernet, enabling message transmission according to a time-triggered (TT) paradigm. A network planning tool is used to allocate each device a finite amount of time in which it may transmit a frame. Each time slot is repeated sequentially to form a periodic communication schedule that is then loaded onto each TTEthernet device (e.g. switches and end systems). Each network participant references the synchronized time in order to dispatch messages at predetermined instances. This schedule guarantees that no contention exists between time-triggered Ethernet frames in the network switches, therefore eliminating the need for arbitration (and the timing variation it causes). Besides time-triggered messaging, TTEthernet networks may provide two additional traffic classes to support communication of different criticality levels. In the rate-constrained (RC) traffic class, the frame payload size and rate of transmission along each communication channel are limited to predetermined maximums. The network switches can therefore be configured to accommodate the known worst-case traffic pattern, and buffer overflows can be eliminated. The best-effort (BE) traffic class behaves akin to classical Ethernet. No guarantees are provided regarding transmission latency or successful message delivery. TTEthernet coordinates transmission of all three traffic classes over the same physical connections, therefore accommodating the full spectrum of traffic criticality levels required in IMA architectures. Common computing platforms (e.g. LRUs) can share networking resources in such a way that failures in non-critical systems (using BE or RC communication modes) cannot impact flight-critical functions (using TT communication). Furthermore, TTEthernet hardware (e.g. switches, cabling) can be shared by both TTEthernet and classical Ethernet traffic.

Loveless, Andrew↗
NASA NTRS1992

A simple, effective media access protocol system for integrated, high data rate networks

The operation and performance of a dual media access protocol for integrated, gigabit networks are described. Unlike other dual protocols, each protocol supports a different class of traffic. The Carrier Sensed Multiple Access-Ring Network (CSMA/RN) protocol and the Circulating Reservation Packet (CRP) protocol support asynchronous and synchronous traffic, respectively. The two protocols operate with minimal impact upon each other. Performance information demonstrates that they support a complete range of integrated traffic loads, do not require call setup/termination or a special node for synchronous traffic control, and provide effective pre-use and recovery. The CRP also provides guaranteed access and fairness control for the asynchronous system. The paper demonstrates that the CSMA-CRP system fulfills many of the requirements for gigabit LAN-MAN networks most effectively and simply. To accomplish this, CSMA-CRP features are compared against similar ring and bus systems, such as Cambridge Fast Ring, Metaring, Cyclic Reservation Multiple Access, and Distributed Dual Queue Data Bus (DQDB).

Foudriat, E. C.↗
NASA NTRS2000

Efficient Use of Distributed Systems for Scientific Applications

Distributed computing has been regarded as the future of high performance computing. Nationwide high speed networks such as vBNS are becoming widely available to interconnect high-speed computers, virtual environments, scientific instruments and large data sets. One of the major issues to be addressed with distributed systems is the development of computational tools that facilitate the efficient execution of parallel applications on such systems. These tools must exploit the heterogeneous resources (networks and compute nodes) in distributed systems. This paper presents a tool, called PART, which addresses this issue for mesh partitioning. PART takes advantage of the following heterogeneous system features: (1) processor speed; (2) number of processors; (3) local network performance; and (4) wide area network performance. Further, different finite element applications under consideration may have different computational complexities, different communication patterns, and different element types, which also must be taken into consideration when partitioning. PART uses parallel simulated annealing to partition the domain, taking into consideration network and processor heterogeneity. The results of using PART for an explicit finite element application executing on two IBM SPs (located at Argonne National Laboratory and the San Diego Supercomputer Center) indicate an increase in efficiency by up to 36% as compared to METIS, a widely used mesh partitioning tool. The input to METIS was modified to take into consideration heterogeneous processor performance; METIS does not take into consideration heterogeneous networks. The execution times for these applications were reduced by up to 30% as compared to METIS. These results are given in Figure 1 for four irregular meshes with number of elements ranging from 30,269 elements for the Barth5 mesh to 11,451 elements for the Barth4 mesh. Future work with PART entails using the tool with an integrated application requiring distributed systems. In particular this application, illustrated in the document entails an integration of finite element and fluid dynamic simulations to address the cooling of turbine blades of a gas turbine engine design. It is not uncommon to encounter high-temperature, film-cooled turbine airfoils with 1,000,000s of degrees of freedom. This results because of the complexity of the various components of the airfoils, requiring fine-grain meshing for accuracy. Additional information is contained in the original.

Taylor, Valerie↗
DOE OSTI2022

Reprocessable polyhydroxyurethane networks reinforced with reactive polyhedral oligomeric silsesquioxanes (POSS) and exhibiting excellent elevated temperature creep resistance

The rapid development of covalent adaptable networks or vitrimers shows promise for addressing the long-standing recycling issues associated with conventional, permanently cross-linked thermosets. At the same time, it is important to demonstrate that properties of reprocessable polymer networks can be optimized to meet the ongoing demand for high-performance materials. We have fabricated reprocessable polyhydroxyurethane (PHU) network composites reinforced with reactive polyhedral oligomeric silsesquioxanes (POSS). With functionalized POSS serving as a fraction of the cross-linkers, the PHU–POSS network nanocomposites exhibit significantly enhanced storage modulus at the rubbery plateau region relative to the neat PHU network. With up to 10 wt% POSS loading, these network composites can undergo melt-state reprocessing at 140 °C with 100% property recovery associated with cross-link density. We also show that hydroxyurethane dynamic chemistry leads to excellent creep resistance at elevated temperature up to 90 °C and is unaffected by reactive incorporation of POSS. In conclusion, this study demonstrates the effectiveness of POSS as nanofillers for designing high-performance, organic-inorganic dynamic PHU networks with excellent reprocessability.

42 ENGINEERING↗
DOE OSTI2019

Software Defined Networking for HPC Interconnects and its Extension across Domains (Final Technical Report for DE-SC0016039)

This project seeks to incorporate OpenFlow-style Software Defined Networking capability into existing High Performance Computing (HPC) interconnect technology, InfiniBand, and to extend such capability beyond one domain. We develop SDN-enhanced InfiniBand that supports OpenFlow functionality for HPC interconnects, investigate the required techniques and the potential benefits for applying SDN on different types of HPC interconnects, and develop software to control the network and expose SDN capability to external networks through the OpenFlow interface. We have also developed techniques to evaluate communication characteristics of parallel applications, which determines the workload on interconnects, investigated newly proposed Interconnection network technologies including new network topologies and routing schemes, and designed novel interconnect routing and modeling techniques. Major activities and findings are summarized as follows.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

HipMCL: a high-performance parallel implementation of the Markov clustering algorithm for large-scale networks

Biological networks capture structural or functional properties of relevant entities such as molecules, proteins or genes. Characteristic examples are gene expression networks or protein–protein interaction networks, which hold information about functional affinities or structural similarities. Such networks have been expanding in size due to increasing scale and abundance of biological data. While various clustering algorithms have been proposed to find highly connected regions, Markov Clustering (MCL) has been one of the most successful approaches to cluster sequence similarity or expression networks. Despite its popularity, MCL’s scalability to cluster large datasets still remains a bottleneck due to high running times and memory demands. In this paper, we present High-performance MCL (HipMCL), a parallel implementation of the original MCL algorithm that can run on distributed-memory computers. We show that HipMCL can efficiently utilize 2000 compute nodes and cluster a network of ~70 million nodes with ~68 billion edges in ~2.4 h. By exploiting distributed-memory environments, HipMCL clusters large-scale networks several orders of magnitude faster than MCL and enables clustering of even bigger networks. Finally, HipMCL is based on MPI and OpenMP and is freely available under a modified BSD license.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1993

High-performance mass storage system for workstations

Reduced Instruction Set Computer (RISC) workstations and Personnel Computers (PC) are very popular tools for office automation, command and control, scientific analysis, database management, and many other applications. However, when using Input/Output (I/O) intensive applications, the RISC workstations and PC's are often overburdened with the tasks of collecting, staging, storing, and distributing data. Also, by using standard high-performance peripherals and storage devices, the I/O function can still be a common bottleneck process. Therefore, the high-performance mass storage system, developed by Loral AeroSys' Independent Research and Development (IR&D) engineers, can offload a RISC workstation of I/O related functions and provide high-performance I/O functions and external interfaces. The high-performance mass storage system has the capabilities to ingest high-speed real-time data, perform signal or image processing, and stage, archive, and distribute the data. This mass storage system uses a hierarchical storage structure, thus reducing the total data storage cost, while maintaining high-I/O performance. The high-performance mass storage system is a network of low-cost parallel processors and storage devices. The nodes in the network have special I/O functions such as: SCSI controller, Ethernet controller, gateway controller, RS232 controller, IEEE488 controller, and digital/analog converter. The nodes are interconnected through high-speed direct memory access links to form a network. The topology of the network is easily reconfigurable to maximize system throughput for various applications. This high-performance mass storage system takes advantage of a 'busless' architecture for maximum expandability. The mass storage system consists of magnetic disks, a WORM optical disk jukebox, and an 8mm helical scan tape to form a hierarchical storage structure. Commonly used files are kept in the magnetic disk for fast retrieval. The optical disks are used as archive media, and the tapes are used as backup media. The storage system is managed by the IEEE mass storage reference model-based UniTree software package. UniTree software will keep track of all files in the system, will automatically migrate the lesser used files to archive media, and will stage the files when needed by the system. The user can access the files without knowledge of their physical location. The high-performance mass storage system developed by Loral AeroSys will significantly boost the system I/O performance and reduce the overall data storage cost. This storage system provides a highly flexible and cost-effective architecture for a variety of applications (e.g., realtime data acquisition with a signal and image processing requirement, long-term data archiving and distribution, and image analysis and enhancement).

Chiang, T.↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2025

Applying Machine Learning and Bayesian Inference to Identify and Locate Moving Anthropogenic Sources Using Distributed Acoustic Sensing Data

Distributed acoustic sensing (DAS) systems, which use existing telecommunication fibers, offer high‐resolution capabilities ideal for recording anthropogenic sources. However, the complexity of urban environments and the large amount of data recorded by DAS require automated methods to efficiently detect and categorize anthropogenic sources. Here, we evaluate how well three machine learning models (k‐nearest neighbor [k‐NN], convolutional neural networks, and recurrent‐convolutional neural networks) can identify various anthropogenic sources recorded by DAS. Our findings reveal that both k‐NN and neural network methods perform well in high signal‐to‐noise ratio (SNR) settings. However, their accuracy decreases at SNRs <4. We also use Kalman filtering, a form of Bayesian inference, on backprojected locations of these sources to recover locations that generally fall within standard smartphone Global Positioning System errors. By combining machine learning and Kalman filter results, we calculate a multidimensional model of moving anthropogenic sources. These results demonstrate the potential of DAS data in urban seismology for accurately identifying and locating such sources. Depending on the research objectives, these sources can be further studied or filtered out to improve the quality of seismic data for earthquake studies. Such methods provide a valuable tool for urban seismology and seismic hazard analysis.

Luckie, Thomas William [Sandia National Laboratori↗
NASA NTRS1995

Criteria development for upgrading computer networks

Being an infrastructure system, the computer network has a fundamental role in the day to day activities of personnel working at KSC. It is easily appreciated that the lack of 'satisfactory' network performance can have a high 'cost' for KSC. Yet, this seemingly obvious concept is quite difficult to demonstrate. At what point do we say that performance is below the lowest tolerable level? How do we know when the 'cost' of using the system at the current level of degraded performance exceeds the cost of upgrading it? In this research, we consider the cost and performance factors that may have an effect in decision making in regards to upgrading computer networks. Cost factors are detailed in terms of 'direct costs' and 'subjective costs'. Performance factors are examined in terms of 'required performance' and 'offered performance.' Required performance is further examined by presenting a methodology for trend analysis based on applying interpolation methods to observed traffic levels. Offered performance levels are analyzed by deriving simple equations to evaluate network performance. The results are evaluated in the light of recommended upgrade policies currently in use for telephone exchange systems, similarities and differences between the two types of services are discussed.

Efe, Kemal↗
DOE OSTI2017

OSiRIS: a distributed Ceph deployment using software defined networking for multi-institutional research

We report on the first year of the OSiRIS project (NSF Award #1541335, UM, IU, MSU and WSU) which is targeting the creation of a distributed Ceph storage infrastructure coupled together with software-defined networking to provide high-performance access for wellconnected locations on any participating campus. The projects goal is to provide a single scalable, distributed storage infrastructure that allows researchers at each campus to read, write, manage and share data directly from their own computing locations. The NSF CC*DNI DIBBS program which funded OSiRIS is seeking solutions to the challenges of multi-institutional collaborations involving large amounts of data and we are exploring the creative use of Ceph and networking to address those challenges. While OSiRIS will eventually be serving a broad range of science domains, its first adopter will be the LHC ATLAS detector project via the ATLAS Great Lakes Tier-2 (AGLT2) jointly located at the University of Michigan and Michigan State University. Part of our presentation will cover how ATLAS is using the OSiRIS infrastructure and our experiences integrating our first user community. The presentation will also review the motivations for and goals of the project, the technical details of the OSiRIS infrastructure, the challenges in providing such an infrastructure, and the technical choices made to address those challenges. We will conclude with our plans for the remaining 4 years of the project and our vision for what we hope to deliver by the projects end.

97 MATHEMATICS AND COMPUTING↗