Search NASA⌕ Search

SEARCH · Search NASA

Results for “storage architectures”

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 379 records · Page 21
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

High Performance Computing Peak Shaving for Microreactor Operation

There are multiple nuclear microreactors currently under development that are designed to provide autonomous power for as many as ten or more years without refueling and are designed to power high performance computing (HPC) datacenters. But the load-follow speeds for a nuclear microreactor will be much slower than grid power and slower than the power variance typical of a HPC system. HPC datacenters experience peak power load variance driven by several factors ranging from the operation of cooling systems to remove heat from the servers to supporting a wide range of user application workflows and architectures each with different power signatures. One mechanism to support the limited load-follow of a microreactor is peak shaving where an energy storage mechanism is used to shed peak load and reduce significant power variance. This work explores peak electrical load shaving using uninterruptible power supply (UPS) systems designed for HPC support in the context of peak shaving when operating using a nuclear microreactor with a load-follow limited to 10% of load per minute. Using a self contained HPC datacenter complete with stand-alone cooling system and provisioned with an x86 cluster, an ARM cluster, and a graphics processing unit (GPU) cluster, peak shaving for microreactor operation using the UPS battery backup is explored while running two classes of typical HPC user applications. HPC architecture suitability for microreactor operation under this type of peak shaving is examined.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

An Architectural Survey of the REECo Maintenance Compound, Area 6, Nevada National Security Site, Nye County, Nevada

The U.S. Department of Energy (DOE), National Nuclear Security Administration Nevada Field Office (NNSA/NFO) plans to install a 26.45-mile-long 138-kilovolt (kV) transmission line on the Nevada National Security Site (NNSS) in Nye County, Nevada. This new transmission line will upgrade the electrical transmission system and provide reliable power between Mercury Switching Station in Area 23 and the U1a Facility in Area 1. The proposed upgrade to the power transmission system constitutes an undertaking subject to review under Section 106 of the National Historic Preservation Act (NHPA) (54 United States Code [USC] § 306101) and its implementing regulations, 36 Code of Federal Regulations (CFR) Part 800. A portion of this new transmission line will cross the western perimeter of the Reynolds Electrical and Engineering Company (REECo) Maintenance Compound located in Area 6 of the NNSS. This compound with its trade shops, storage sheds, and yards served as a general construction facility that supported nuclear testing activities on the NNSS from 1967 to 1992. Although three architectural resources in the compound have previously been recorded and evaluated for the National Register of Historic Places (NRHP), other resources within it have not, nor has the compound been recorded and evaluated as a potential district. The State Historic Preservation Officer (SHPO) concurred with NNSA/NFO’s determination that the unevaluated REECo Maintenance Compound will be adversely affected by the proposed undertaking (Reed 2021). The NNSA/NFO, in consultation with the SHPO, then developed the Memorandum of Agreement DE-GM58-22NA25553 between the U.S. Department of Energy and the Nevada State Historic Preservation Officer Regarding Installation of a 138-kilovolt Transmission Line from the Mercury Switching Station to the U1a Facility and the Removal of the Historic 138-kilovolt Transmission Line from the Mercury Switching Station to the U1a Facility in Areas 1, 3, 5, 6, and 23 of the Nevada National Security Site, Nye County, hereafter referred to as the MOA. This architectural survey report was prepared in accordance with MOA Stipulation III.B. It describes the origins, history, layout, and functions of the REECo Maintenance Compound during the Cold War and identifies contributing and noncontributing resources. The report defines a district boundary based on archival research and the field survey and concludes with an appendix that presents digital color images of the compound with descriptions and a map key showing image viewpoints and directions. Although none of the architectural resources are recommended individually eligible, the district evaluation concludes the REECo Maintenance Compound is eligible as a historic district for listing in the NRHP under the Secretary of the Interior’s Significance Criteria A and C at the local level. The historic district retains all seven aspects of integrity and conveys its significance under the abovementioned criteria.

138 kV↗
NASA NTRS1995

Nuclear Thermal Rocket/Vehicle Design Options for Future NASA Missions to the Moon and Mars

The nuclear thermal rocket (NTR) provides a unique propulsion capability to planners/designers of future human exploration missions to the Moon and Mars. In addition to its high specific impulse (approximately 850-1000 s) and engine thrust-to-weight ratio (approximately 3-10), the NTR can also be configured as a 'dual mode' system capable of generating electrical power for spacecraft environmental systems, communications, and enhanced stage operations (e.g., refrigeration for long-term liquid hydrogen storage). At present the Nuclear Propulsion Office (NPO) is examining a variety of mission applications for the NTR ranging from an expendable, single-burn, trans-lunar injection (TLI) stage for NASA's First Lunar Outpost (FLO) mission to all propulsive, multiburn, NTR-powered spacecraft supporting a 'split cargo-piloted sprint' Mars mission architecture. Each application results in a particular set of requirements in areas such as the number of engines and their respective thrust levels, restart capability, fuel operating temperature and lifetime, cryofluid storage, and stage size. Two solid core NTR concepts are examined -- one based on NERVA (Nuclear Engine for Rocket Vehicle Application) derivative reactor (NDR) technology, and a second concept which utilizes a ternary carbide 'twisted ribbon' fuel form developed by the Commonwealth of Independent States (CIS). The NDR and CIS concepts have an established technology database involving significant nuclear testing at or near representative operating conditions. Integrated systems and mission studies indicate that clusters of two to four 15 to 25 klbf NDR or CIS engines are sufficient for most of the lunar and Mars mission scenarios currently under consideration. This paper provides descriptions and performance characteristics for the NDR and CIS concepts, summarizes NASA's First Lunar Outpost and Mars mission scenarios, and describes characteristics for representative cargo and piloted vehicles compatible with a reference 240 t-class heavy lift launch vehicle (HLLV) and smaller 120 t HLLV option. Attractive performance characteristics and high-leverage technologies associated with both the engine and stage are identified, and supporting parametric sensitivity data is provided. The potential for commonality of engine and stage components to satisfy a broad range of lunar and Mars missions is also discussed.

Borowski, Stanley K.↗
NASA NTRS1987

The Penn State astronomical image processing system

The needs of modern astronomy for image processing set demanding standards in simultaneously requiring fast computation speed, high-quality graphic display, large data storage, and interactive response. An innovative image processing system was designed, integrated, and used; it is based on a supermicro architecture which is tailored specifically for astronomy, which provides a highly cost-effective alternative to the traditional minicomputer installation. The paper describes the design rationale, equipment selection, and software developed to allow other astronomers with similar needs to benefit from the present experience.

Truax, Ryland J.↗
NASA NTRS1998

Sixth Goddard Conference on Mass Storage Systems and Technologies Held in Cooperation with the Fifteenth IEEE Symposium on Mass Storage Systems

This document contains copies of those technical papers received in time for publication prior to the Sixth Goddard Conference on Mass Storage Systems and Technologies which is being held in cooperation with the Fifteenth IEEE Symposium on Mass Storage Systems at the University of Maryland-University College Inn and Conference Center March 23-26, 1998. As one of an ongoing series, this Conference continues to provide a forum for discussion of issues relevant to the management of large volumes of data. The Conference encourages all interested organizations to discuss long term mass storage requirements and experiences in fielding solutions. Emphasis is on current and future practical solutions addressing issues in data management, storage systems and media, data acquisition, long term retention of data, and data distribution. This year's discussion topics include architecture, tape optimization, new technology, performance, standards, site reports, vendor solutions. Tutorials will be available on shared file systems, file system backups, data mining, and the dynamics of obsolescence.

Kobler, Benjamin↗
NASA NTRS2017

Regenerative Fuel Cell Power Systems for Lunar and Martian Surface Exploration

This paper presents the preliminary results of a recent National Aeronautics and Space Administration (NASA) study funded under the Advanced Exploration Systems (AES) Modular Power Systems (AMPS) project. This study evaluated multiple surface locations on both the Moon and Mars, with the goal of establishing a common approach towards technology development and system design for surface power systems that use Regenerative Fuel Cell (RFC) energy storage methods. One RFC design may not be applicable to all surface locations; however, AMPS seeks to find a unified architecture, or series of architectures, that leverages a single development approach to answer the technology need for RFC systems. Early system trades were performed to select the most effective fuel cell and electrolyzer architectures based on current state-of-the-art technology, whereas later trades will establish a detailed system design to enable a near-term ground (non-flight) demonstration. This paper focuses on the initial trade studies, presents the selected fuel cell and electrolyzer architectures for follow-on system design studies, and suggests areas for further technology investment.

lunar and Mars exploration↗
NASA NTRSDate not supplied

200 Gbps TBIRD CubeSat Downlink: Pre-Flight Test Results

The Terabyte Infrared Delivery (TBIRD) program will establish an optical communication link from a 6Unanosatellite in low-Earth orbit to a ground station at burst rates up to 200 Gbps. The system is capable of reliable data delivery from a 2 TB storage buffer on the payload to a ground terminal in the presence of atmospheric fading. An overview of the communication architecture for TBIRD is provided as well as results from communications performance testing of the 3U lasercom payload prior to spacecraft integration. Launch is scheduled for mid-year 2022.

space communications↗
DOE OSTI2022

Next Generation System Analysis Model Recently Added Features and Future Plans - Abstract

The Nuclear Waste Policy Act of 1982, as amended (NWPA 1982), established the federal government’s responsibility to accept spent nuclear fuel (SNF) and high-level radioactive waste (HLW) from waste owners and generators for ultimate disposition. SNF generated by the current fleet of commercial nuclear reactors is being stored at the reactor sites in spent fuel pools (SFPs) and in dry independent spent fuel storage installations (ISFSIs). The US Department of Energy Office of Nuclear Energy (DOE-NE) is developing an Integrated Waste Management Program (IWMP) comprising a suite of options and supporting analyses to enable future informed choices. The IWMP is applying integrated waste management system architecture analysis, system engineering, and decision analysis principles to inform potential future decisions regarding potential nuclear waste management system architectures. Architecture analyses of the IWM system are being conducted to support the future deployment of a comprehensive system for managing nuclear waste that considers all major aspects of the back end of the nuclear fuel cycle (i.e., transportation, storage, and disposal). The Next Generation System Analysis Model (NGSAM) is an agent-based simulation software tool designed for the express purpose of modeling the IWM system. NGSAM imports data from the Oak Ridge National Laboratory (ORNL) Unified Database (e.g., historic assembly information, thermal profiles for assembly heat, at-reactor dry storage loadings) to ensure that the simulation initializes with a realistic representation of the state of commercial SNF in the United States. Recent major enhancements that have been implemented into NGSAM since NGSAM was last presented at the WM2019 conference include: • Tracking of railroad escort and buffer car acquisition. • Addition of heavy haul and barge routes for some sites, as well as support for user-defined inter-modal routes. • Updates to the logic that checks the thermal maps prior to package transport. • Addition of an allocation method that predicts when reactor sites will pack assemblies from their pools for dry storage and allocates packages to those reactor sites in the preceding periods, favoring direct transport packages and reducing the number of packages that reactor sites pack for dry storage at their ISFSIs. • Addition of reactor site family operational limits, which are used to limit the number of loads from the pool and from dry storage at a given reactor site per year. • Support has been added for multiple canister loading maps and packages having multiple compatible transportation overpacks. • Updates in the handling of non-commercial fuel, including a new database containing data to support the updates. • Support for repackaging at reactor sites. • Implementing additional output reports or modifying existing reports. • User edits can now be created and edited via the NGSAM website. • Ability to load packages for dry storage at ISF pools. • Same-type package blending at DOE sites. • Support for multi-mode transloading at reactor sites. These new features have improved NGSAM capabilities and/or improve the user experience with the model and will be discussed in more detail. The initial NGSAM requirements for advanced reactor fuels, reprocessing, treatment, and conditioning are preliminary and are described at a high level in this paper: analysts will provide more specific requirements to the NGSAM team in the future. Additionally, there are many data needs associated with modeling advanced reactors in NGSAM, but many of the data or plans are still in progress and/or yet to be fully defined. However, this document describes an initial exploration of the data relevant to this program. Advanced reactor data will likely require revision as concepts evolve and new considerations are made. This is a technical paper that does not take into account contractual limitations or obligations under the Standard Contract for Disposal of Spent Nuclear Fuel and/or High-Level Radioactive Waste (Standard Contract) (10 CFR Part 961). For example, under the provisions of the Standard Contract, spent nuclear fuel in multi-assembly canisters is not an acceptable waste form, absent a mutually agreed to contract amendment. To the extent discussions or recommendations in this paper conflict with the provisions of the Standard Contract, the Standard Contract governs the obligations of the parties, and this paper in no manner supersedes, overrides, or amends the Standard Contract. This paper reflects technical work which could support future decision making by DOE. No inferences should be drawn from this paper regarding future actions by DOE, which are limited both by the terms of the Standard Contract and Congressional appropriations for the Department to fulfill its obligations under the Nuclear Waste Policy Act including licensing and construction of a spent nuclear fuel repository.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Lustre Unveiled: Evolution, Design, Advancements, and Current Trends

The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Three-Dimensional Microbatteries beyond Lithium Ion

Ubiquitous implementation of microelectronics in microelectromechanical systems requires stable power supplies and necessitates development of on-chip miniature power sources, such as microbatteries. However, conventional thin-film batteries exhibit limited surface area and sluggish diffusion kinetics with insufficient energy and power density. In contrast, three-dimensional (3D) beyond-lithium (sodium, zinc, aluminum, etc.) battery architectures can significantly enhance the areal energy and power and meanwhile maintain the low-cost mass production. Despite this, the future of beyond-lithium systems is being questioned as they each present shortcomings. In this Comment, we discuss scientific advancements in reaction kinetics control, electrochemical stability improvements, and reaction mechanism understandings in 3D beyond-lithium battery systems, and finally, call for intensive research effort to address the challenges and develop strategies that can pave the way toward their commercialization.

25 ENERGY STORAGE↗
NASA NTRS2014

Technology Maturation in Preparation for the Cryogenic Propellant Storage and Transfer (CPST) Technology Demonstration Mission (TDM)

In support of its goal to find an innovative path for human space exploration, NASA embarked on the Cryogenic Propellant Storage and Transfer (CPST) Project, a Technology Demonstration Mission (TDM) to test and validate key cryogenic capabilities and technologies required for future exploration elements, opening up the architecture for large in-space cryogenic propulsion stages and propellant depots. Recognizing that key Cryogenic Fluid Management (CFM) technologies anticipated for on-orbit (flight) demonstration would benefit from additional maturation to a readiness level appropriate for infusion into the design of the flight demonstration, the NASA Headquarters Space Technology Mission Directorate (STMD) authorized funding for a one-year technology maturation phase of the CPST project. The strategy, proposed by the CPST Project Manager, focused on maturation through modeling, concept studies, and ground tests of the storage and fluid transfer of CFM technology sub-elements and components that were lower than a Technology Readiness Level (TRL) of 5. A technology maturation plan (TMP) was subsequently approved which described: the CFM technologies selected for maturation, the ground testing approach to be used, quantified success criteria of the technologies, hardware and data deliverables, and a deliverable to provide an assessment of the technology readiness after completion of the test, study or modeling activity. The specific technologies selected were grouped into five major categories: thick multilayer insulation, tank applied active thermal control, cryogenic fluid transfer, propellant gauging, and analytical tool development. Based on the success of the technology maturation efforts, the CPST project was approved to proceed to flight system development.

propellants↗
DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multi-threaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0x and 6.8x on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3x over the multi-thread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗
DOE OSTI2021

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗
DOE OSTI2020

Integrating System to Edge-of-Network Architecture and Management for SHINES (SEAMS) Technologies of High Penetration Grids

Consistent with the U.S. DOE’s EERE SHINES FOA objectives, the goal of this project was to enable integration from EMS to variable DER in a way that delivered visibility and opportunities for managing and controlling distributed resources using SEAMS technologies. This project implemented several SHINES technologies, which included: (1) three Stem battery systems; (2) six ConnectDER PV collars; (3) two E-Gear PV/Battery storage interfaces; and (4) eight Kitu smart inverter emulators. A Siemens Energy IP DEMS was deployed in a development environment and integrated with the SHINES technologies via OpenADR and IEEE 2030.5 protocols with documented architecture. In the case of IEEE 2030.5, a suite of technologies including a gateway, an API, software and a server from Kitu Systems (the Citadel, Convoy, Kitu Crosslink API, and Kitu Spark) were used to demonstrate that the utility could achieve visibility of distributed energy resources (DER) and control of single and multiple DER systems with a single command. The project also included a LCOE analysis of SHINES technologies compared to traditional system and circuit mitigation measures.

14 SOLAR ENERGY↗
NASA NTRSDate not supplied

Commercially-viable Hydrogen Aircraft for Reduction of Greenhouse Emissions

NASA assembled a cross-organizational multidisciplinary “radical” project team combining a diverse set of skills including aircraft architecture modelling and optimization, advanced material science, and engineering of high performance cryogenic, thermal management and fuel cell systems components and subsystems to tackle the challenging problem of development of commercially viable mid-size aircraft that would radically transform air transportation. Our team is developing an integrated conceptual and experimental methodology to realize a medium-range hydrogen aircraft design based on fuel cells, hydrogen burning engines, advanced power management and distribution, cryogenic hydrogen storage systems, and novel thermal management systems combined with an integrated aircraft concept of operations both during the flight and at the airports. The resulting analyses suggested the aircraft architecture options, sizes and layouts for the propulsion subsystem and cryogenic liquid hydrogen (LH2) tankage to verify the weight-scaling relationships for a medium-range aircraft carrying 100-200 passengers flying 1000 - 5000 km. Hydrogen - based distributed electric propulsion and cryogenic systems were further analysed, and more detailed study identified systems goals for a viable overall system weight for missions of various lengths. The developed aircraft architecture is being optimized by total specific energy density, specific power, size and mission profiles.

aircraft architecture↗
NASA NTRSDate not supplied

Commercially-viable Hydrogen Aircraft for Reduction of Greenhouse Emissions

NASA assembled a cross-organizational multidisciplinary “radical” project team combining a diverse set of skills including aircraft architecture modelling and optimization, advanced material science, and engineering of high performance cryogenic, thermal management and fuel cell systems components and subsystems to tackle the challenging problem of development of commercially viable mid-size aircraft that would radically transform air transportation. Our team is developing an integrated conceptual and experimental methodology to realize a medium-range hydrogen aircraft design based on fuel cells, hydrogen burning engines, advanced power management and distribution, cryogenic hydrogen storage systems, and novel thermal management systems combined with an integrated aircraft concept of operations both during the flight and at the airports. The resulting analyses suggested the aircraft architecture options, sizes and layouts for the propulsion subsystem and cryogenic liquid hydrogen (LH2) tankage to verify the weight-scaling relationships for a medium-range aircraft carrying 100-200 passengers flying 1000 - 5000 km. Hydrogen - based distributed electric propulsion and cryogenic systems were further analysed, and more detailed study identified systems goals for a viable overall system weight for missions of various lengths. The developed aircraft architecture is being optimized by total specific energy density, specific power, size and mission profiles.

aircraft architecture↗
DOE OSTI2021

A High Performance Sparse Tensor Algebra Compiler in MLIR

Sparse tensor algebra is widely used in many applications, including scientific computing, machine learning, and data analytics. The performance of sparse tensor algebra kernels strongly depends on the intrinsic characteristics of the input tensors, hence many storage formats are designed for tensors to achieve optimal performance for particular applications/architectures, which makes it challenging to implement and optimize every tensor operation of interest on a given architecture. We propose a tensor algebra domain-specific language (DSL) and compiler framework to automatically generate kernels for mixed sparse-dense tensor algebra operations. The proposed DSL provides high-level programming abstractions that resemble the familiar Einstein notation to represent tensor algebra operations. The compiler introduces a new Sparse Tensor Algebra dialect built on top of LLVM's extensible MLIR compiler infrastructure for efficient code generation while covering a wide range of tensor storage formats. Our compiler also leverages input-dependent code optimization to enhance data locality for better performance. Our results show that the performance of automatically generated kernels outperforms the state-of-the-art sparse tensor algebra compiler, with up to 20.92x, 6.39x, and 13.9x performance improvement over state-of-the-art tensor algebra compilers, for parallel SpMV, SpMM, and TTM, respectively.

Tian, Ruiqin↗