The Computing Architectures Donnybrook
Explains the context and framework I'm creating as the Computer Architectures associate editor for IEEE Computer Magazine. The article is largely a historical recitation of computer architectures.
SEARCH · Search NASA
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.
Explains the context and framework I'm creating as the Computer Architectures associate editor for IEEE Computer Magazine. The article is largely a historical recitation of computer architectures.
Execution on massively threaded processors is one of the most critical challenges for high-performance computing (HPC) scientific visualization. Of the many fundamental changes we are seeing in HPC systems, one of the most profound is a reliance on new processor types optimized for execution bandwidth over latency hiding. Moreover, our current production scientific visualization software is not designed for these new types of architectures. In order to address this issue, the VTK-m framework serves as a container for algorithms, provides flexible data representation, and simplifies the design of visualization algorithms on new and future computer architecture.
Reliability, selectivity, and sensitivity are the fundamental attributes of any protection system, acting as the main drivers in the selection of schemes, and equipment. In high-voltage systems, microprocessor-based relays represent the industry’s preferred solution, providing engineers with a vast array of benefits. However, they remain vulnerable to cybersecurity events that may compromise their functionality. To help mitigate against potential cybersecurity risks, this paper presents a fault-tolerant, Byzantine Resilient (BR) architecture that significantly increases the cybersecurity attributes of a protection system while minimizing the amount of performance impacts and integration overheads introduced. The solution relies on an array of independent relays that utilize robust consensus methods (based on Spire [1], [2]) to ensure correct system behavior is achieved even when a relay has been compromised. Furthermore, the solution has been complemented with a custom-built Situational Awareness engine that can be used to detect and identify potential threats. The implemented solution has been developed in consultation with three hardware vendors and has been tested to comply with the performance requirements of a 345kV differential protection scheme (87T). The results indicate that the proposed architecture is a comprehensive solution that: supports the strict correctness and performance requirements of the bulk power grid while providing a cost-effective alternative that offers a seamless, long-term solution.
The power transferred through high voltage direct current transmission (HVdc) systems is increasing with increased resilience and reliability needs as well as large scale deployment of renewable energy systems. In this paper, a new HVdc substation (SS) architecture with higher power transfer capability is proposed. The HVdc converters or valve groups (VG) are connected in series and parallel fashion such that the total power rating of the SS is increased. The control challenges associated with such SS architecture are identified and appropriate control modifications are proposed. The same is verified through PSCAD based simulations.
Optimal operations of distribution systems to enhance reliability and resiliency can be achieved if information with higher fidelity can be accessed from grid premise devices. This paper discusses a hierarchical control architecture that can introduce higher visibility and controllability by utilizing the estimated injection and absorption of different behind-the-meter controllable loads, and DERs, including EVs, within a premise. The acquired information is used to make control decisions for scheduling the premise appliances to reduce energy consumptions. The premise controller can also optimize the consumptions on a day-to-day basis in addition to taking input signals from the utility operators to support grid-service events (e.g., peak shaving). The entire control architecture is designed in a modular and scalable fashion that provides benefits to the utility from the reuse of existing infrastructure while creating opportunities for various grid services to be enabled by the operation of premise controllers.
For high-performance computing (HPC) system designers and users, meeting the myriad challenges of next-generation exascale supercomputing systems requires rethinking their approach to application and system software design. Among these challenges, providing resiliency and stability to the scientific applications in the presence of high fault rates requires new approaches to software architecture and design. As HPC systems become increasingly complex, they require intricate solutions for detection and mitigation for various modes of faults and errors that occur in these large-scale systems, as well as solutions for failure recovery. These resiliency solutions often interact with and affect other system properties, including application scalability, power and energy efficiency. Therefore, resilience solutions for HPC systems must be thoughtfully engineered and deployed.In previous work, we developed the concept of resilience design patterns, which consist of templated solutions based on well-established techniques for detection, mitigation and recovery. In this paper, we use these patterns as the foundation to propose new approaches to designing runtime systems for HPC systems. The instantiation of these patterns within a runtime system enables flexible and adaptable end-to-end resiliency solutions for HPC environments. The paper describes the architecture of the runtime system, named Plexus, and the strategies for dynamically composing and adapting pattern instances under runtime control. This runtime-based approach enables actively balancing the cost-benefit trade-off between performance overhead and protection coverage of the resilience solutions. Based on a prototype implementation of PLEXUS, we demonstrate the resiliency and performance gains achieved by the pattern-based runtime system for a parallel linear solver application.
The nuclear sector is pursuing a number of advanced reactor concepts, one of which is the microreactor, an advanced reactor characterized by a power output of less than 20 MWth. These reactors are intended for use in applications where traditional power solutions are economically or logistically impractical. One key feature required for the successful deployment of microreactors is a remote operation capability, which can dramatically cut staffing expenses by eliminating the need for licensed operators to be present at the site of each reactor and instead concentrate in a centrally located operations center. In moving to a remote operations framework for nuclear reactors, the number of potential attack surfaces for a cyber adversary looking to cause harm or disruption increases. Therefore, a robust cybersecurity architecture is required to mitigate these potential vulnerabilities. This paper explores the functional infrastructure, security, and communication requirements to adapt remote operations to nuclear applications. It then presents a reference architecture for remote operations of microreactors that applies best practices in cybersecurity and remote communications in the context of a digital twin remote operation system.
Triadic analysis encompasses a useful set of graph mining methods that is centered on the concept of a triad, which is a subgraph of three nodes and the configuration of directed edges across the nodes. Such methods are often applied in the social sciences as well as many other diverse fields. Triadic methods commonly operate on a triad census that counts the number of triads of every possible edge configuration in a graph. Like other graph algorithms, triadic census algorithms do not scale well when graphs reach tens of millions to billions of nodes. To enable the triadic analysis of large-scale graphs, we developed and optimized a triad census algorithm to efficiently execute on shared memory architectures. We will retrace the development and evolution of a parallel triad census algorithm. Over the course of several versions, we continually adapted the code’s data structures and program logic to expose more opportunities to exploit parallelism on shared memory that would translate into improved computational performance. We will recall the critical steps and modifications that occurred during code development and optimization. Furthermore, we will compare the performances of triad census algorithm versions on three specific systems: Cray XMT, HP Superdome, and AMD multi-core NUMA machine. These three systems have shared memory architectures but with markedly different hardware capabilities to manage parallelism.
e present an efficient implementation of the radix sort algo- rithm for the Tilera TILEPro64 processor. The TILEPro64 is one of the first successful commercial manycore processors. It is com- posed of 64 tiles interconnected through multiple fast Networks- on-chip and features a fully coherent, shared distributed cache. The architecture has a large degree of flexibility, and allows various optimization strategies. We describe how we mapped the algorithm to this architecture. We present an in-depth analysis of the optimizations for each phase of the algorithm with respect to the processor’s sustained performance. We discuss the overall throughput reached by our radix sort implementation (up to 132 MK/s) and show that it provides comparable or better performance-per-watt with respect to state-of-the art implemen- tations on x86 processors and graphic processing units.
Understanding the most efficient design and utilization of emerging multicore systems is one of the most challenging questions faced by the mainstream and scientific computing industries in several decades. Our work explores multicore stencil (nearest-neighbor) computations — a class of algorithms at the heart of many structured grid codes, including PDE solvers. We develop a number of effective optimization strategies, and build an auto-tuning environment that searches over our optimizations and their parameters to minimize runtime, while maximizing performance portability. To evaluate the effectiveness of these strategies we explore the broadest set of multicore architectures in the current HPC literature, including the Intel Clovertown, AMD Barcelona, Sun Victoria Falls, IBM QS22 PowerXCell 8i, and NVIDIA GTX280. Overall, our auto-tuning optimization methodology results in the fastest multicore stencil performance to date. Finally, we present several key insights into the architectural tradeoffs of emerging multicore designs and their implications on scientific algorithm development.
This study presents a photovoltaic (PV) inverter architecture composed of stackable dc to three-phase ac converter blocks. Several such blocks, each containing a converter power stage and controls, are connected in series on their ac sides to obtain transformerless medium-voltage ac interfaces for PV power plants. The series-connected structure is made possible by a quadruple active bridge dc-dc converter that provides isolation between the PV input and each of the three ac-side phases within each block. Furthermore, since incoming PV power is transferred as constant balanced three-phase ac power, instantaneous inputoutput power balance bypasses the need for bulk energy storage. To streamline implementation and maximize system scalability and resilience, decentralized block-level controllers accomplish dc-link voltage regulation, maximum power point tracking, and ac-side power sharing without centralized means. The proposed architecture is validated by simulations of a PV string to mediumvoltage ac system consisting of six blocks and on a proof-of-concept hardware prototype that consists of three cascaded converter blocks.
OpenPET is an open source, modular, extendible, and high-performance platform suitable for multi-channel data acquisition and analysis. Due to the versatility of the hardware, firmware, and software architectures, the platform is capable of interfacing with a wide variety of detector modules not only in medical imaging but also in homeland security applications. Analog signals from radiation detectors share similar characteristics-a pulse whose area is proportional to the deposited energy and whose leading edge is used to extract a timing signal. As a result, a generic design method of the platform is adopted for the hardware, firmware, and software architectures and implementations. The analog front-end is hosted on a module called a Detector Board, where each board can filter, combine, timestamp, and process multiple channels independently. The processed data is formatted and sent through a backplane bus to a module called Support Board, where 1 Support Board can host up to eight Detector Board modules. The data in the Support Board, coming from 8 Detector Board modules, can be aggregated or correlated (if needed) depending on the algorithm implemented or runtime mode selected. It is then sent out to a computer workstation for further processing. The number of channels (detector modules), to be processed, mandates the overall OpenPET System Configuration, which is designed to handle up to 1,024 channels using 16-channel Detector Boards in the Standard System Configuration and 16,384 channels using 32-channel Detector Boards in the Large System Configuration.
Recent trends of CMOS scaling and increasing number of on-chip cores have led to a large increase in the size of on-chip caches. Since SRAM has low density and consumes large amount of leakage power, its use in designing on-chip caches has become more challenging. To address this issue, researchers are exploring the use of several emerging memory technologies, such as embedded DRAM, spin transfer torque RAM, resistive RAM, phase change RAM and domain wall memory. In this paper, we survey the architectural approaches proposed for designing memory systems and, specifically, caches with these emerging memory technologies. To highlight their similarities and differences, we present a classification of these technologies and architectural approaches based on their key characteristics. We also briefly summarize the challenges in using these technologies for architecting caches. We believe that this survey will help the readers gain insights into the emerging memory device technologies, and their potential use in designing future computing systems.
In this paper, a novel distributed hierarchical coordinated control architecture is proposed for large scale power systems. The newly considered architecture facilitates frequency restoration and power balancing functions to be decoupled and implemented at different levels. At the local level, decentralized robust generator controllers are designed to quickly restore frequency after large faults and disturbances in the system. The controllers presented herein are shown to improve transient stability performance, as compared to conventional governor and excitation control. At the area level, Automatic Generation Control (AGC) is modified and coordinates with the decentralized robust controllers to reach the interchange schedule in the tie lines. The interaction of local and zonal controllers is validated through detailed simulations.
Abstract The performance of braided ceramic matrix composites has been shown to depend on the spatial arrangement of tows; therefore, a new class of tools is required to measure irregularities in the composite architecture for components with intricate geometries. We report a scalable and robust reconstruction technique built upon stereoscopic digital image correlation that is able to efficiently measure the position of tows in arbitrarily shaped composites. This method was applied to triaxially braided ceramic matrix composite tubes intended for use as nuclear fuel cladding, which revealed both long‐range “systematic” tow packing defects associated with the manufacturing process and short‐range “intrinsic” defects due to the braid architecture. These findings suggested that the character of tow spacing variation in braided composite tubes was substantially more complex than in planar woven composites. These measurements are expected to lead to improved processing of braided composites and to facilitate the design of statistically representative virtual specimens for finite element modeling.
Terpenoid-based essential oils are economically important commodities, yet beyond their biosynthetic pathways, little is known about the genetic architecture of terpene oil yield from plants. Transport, storage, evaporative loss, transcriptional regulation and precursor competition may be important contributors to this complex trait. Here, we associate 2.39 million single nucleotide polymorphisms derived from shallow whole-genome sequencing of 468 Eucalyptus polybractea individuals with 12 traits related to the overall terpene yield, eight direct measures of terpene concentration and four biomass-related traits. Our results show that in addition to terpene biosynthesis, development of secretory cavities, where terpenes are both synthesized and stored, and transport of terpenes were important components of terpene yield. For sesquiterpene concentrations, the availability of precursors in the cytosol was important. Candidate terpene synthase genes for the production of 1,8-cineole and α-pinene, and β-pinene (which comprised > 80% of the total terpenes) were functionally characterized as a 1,8-cineole synthase and a β/α-pinene synthase. Our results provide novel insights into the genomic architecture of terpene yield and we provide candidate genes for breeding or engineering of crops for biofuels or the production of industrially valuable terpenes.
Undoped and nitrogen doped TiO{sub 2} thin films were deposited by atomic layer deposition on planar substrates. Deposition on 3D-architecture substrates made of metallic foams was also investigated to propose architectured photovoltaic stack fabrication. All the films were deposited at 265 °C and nitrogen incorporation was achieved by using titanium isopropoxide, NH{sub 3} and/or N{sub 2}O as precursors. The maximum nitrogen incorporation level obtained in this study was 2.9 at. %, resulting in films exhibiting a resistivity of 115 Ω cm (+/−10 Ω cm) combined with an average total transmittance of 60% in the 400–1000 nm wavelength range. Eventually, TiO{sub 2} thin films were deposited on the 3D metallic foam template.
The high spectral resolution afforded by Hyperspectral Imaging (HSI) sensors is poised to bring unprecedented advancements to signature characterization applications. Thus far, much of the research in the machine learning field devoted to HSI applications has focused on a few specific tasks like land-use land-cover classification. In land classification tasks, spatial information is very important, and model architectures are often designed to leverage spatial contexts. However, it is unclear how well these spatially-tuned models will translate to tasks where spectral information is critical, like the detection and characterization of chemicals. In this work, we compare spectral models (inputs are 1D spectra) and spatial-spectral models (inputs are 3D cubes) in the context of predicting chemical concentration maps. We find that spatial-spectral models perform the best, though we find a wide range in performance across the different architectures tested. Additionally, we find that model performance is impacted by the availability of training data, particularly in scenarios where the training data doesn't fully capture the true variance of real-world conditions. We find that data augmentation can help mitigate sparse coverage of observed parameter space (e.g., seasonal or geographic variability in ground cover), and present augmentation strategies that are tailored to hyperspectral data.