skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Tensor Algebra Library for NVidia Graphics Processing Units

Abstract

This is a general purpose math library implementing basic tensor algebra operations on NVidia GPU accelerators. This software is a tensor algebra library that can perform basic tensor algebra operations, including tensor contractions, tensor products, tensor additions, etc., on NVidia GPU accelerators, asynchronously with respect to the CPU host. It supports a simultaneous use of multiple NVidia GPUs. Each asynchronous API function returns a handle which can later be used for querying the completion of the corresponding tensor algebra operation on a specific GPU. The tensors participating in a particular tensor operation are assumed to be stored in local RAM of a node or GPU RAM. The main research area where this library can be utilized is the quantum many-body theory (e.g., in electronic structure theory).

Authors:
Publication Date:
Research Org.:
Oak Ridge National Laboratory
Sponsoring Org.:
USDOE
OSTI Identifier:
1253357
Report Number(s):
NV-TAL; 003408MLTPL00
DOE Contract Number:
AC05-00OR22725
Resource Type:
Software
Software Revision:
00
Software Package Number:
003408
Software Package Contents:
Open Source Software package available from Oak Ridge National Laboratory at the following URL: https://github.com/DmitryLyakh/TAL_SH
Software CPU:
MLTPL
Open Source:
Yes
Source Code Available:
No
Other Software Info:
This open source software module (library) has been incorporated into an open source software package ACES IV developed at the University of Florida.
Country of Publication:
United States

Citation Formats

Liakh, Dmitry. Tensor Algebra Library for NVidia Graphics Processing Units. Computer software. https://www.osti.gov//servlets/purl/1253357. Vers. 00. USDOE. 16 Mar. 2015. Web.
Liakh, Dmitry. (2015, March 16). Tensor Algebra Library for NVidia Graphics Processing Units (Version 00) [Computer software]. https://www.osti.gov//servlets/purl/1253357.
Liakh, Dmitry. Tensor Algebra Library for NVidia Graphics Processing Units. Computer software. Version 00. March 16, 2015. https://www.osti.gov//servlets/purl/1253357.
@misc{osti_1253357,
title = {Tensor Algebra Library for NVidia Graphics Processing Units, Version 00},
author = {Liakh, Dmitry},
abstractNote = {This is a general purpose math library implementing basic tensor algebra operations on NVidia GPU accelerators. This software is a tensor algebra library that can perform basic tensor algebra operations, including tensor contractions, tensor products, tensor additions, etc., on NVidia GPU accelerators, asynchronously with respect to the CPU host. It supports a simultaneous use of multiple NVidia GPUs. Each asynchronous API function returns a handle which can later be used for querying the completion of the corresponding tensor algebra operation on a specific GPU. The tensors participating in a particular tensor operation are assumed to be stored in local RAM of a node or GPU RAM. The main research area where this library can be utilized is the quantum many-body theory (e.g., in electronic structure theory).},
url = {https://www.osti.gov//servlets/purl/1253357},
doi = {},
year = 2015,
month = 3,
note =
}

Software:
To order this software, request consultation services, or receive further information, please fill out the following request.

Save / Share:
  • This report documents the idiosyncrasies of the Tektronix PLOT 10 Terminal Control System (TCS) level 3.3 software as it currently exists on the Oak Ridge National Laboratory Fusion Energy Division DECsystem-10 computer. It pertains to the graphics software library called TCSTEK, which may be used to convert TCS subroutine calls into display images on the screens of Tektronix 4000 series storage tube terminals. Fully described here are several user-oriented enhancements and system efficiency modifications to the vendor-supplied TCS software library. It is intended that this report serve as a reference for future TCS updates so that continuity between releases ofmore » TCS PLOT 10 software may be maintained. 1 figure.« less
  • This report documents the user-level subroutines of the TCSTKF software library for the Oak Ridge National Laboratory (ORNL) Fusion Energy Division (FED) DECsystem-10. The TCSTKF graphics library was written and is maintained so that large-production computer programs can access a small, efficient graphics library and produce device-independent graphics files. This library is presented as an alternative to the larger graphics software libraries, such as DISSPLA. The main external difference between this software and the TCSTEK software library is that the TCSTKF software will created .TKF formatted intermediate plot data files, as well as producing display images on the screen ofmore » a Tektronix 4000 series storage tube terminal. These intermediate plot data files can be subsequently postprocessed into report-quality images on a variety of other graphics devices at ORNL.« less
  • This document describes a set of FORTRAN subroutines collectively called the CALVEC subprogram library. The purpose of the CALVEC software library is the emulation of CALCOMP pen and ink graphics on a DECsystem 10. A user level interface with CALVEC software allows standard CALCOMP subprogram calls to produce a VECtor file, SEGMNT.VEC. This vector file may subsequently be postprocessed into an image in a variety of ways.
  • this document describes a set of FORTRAN subroutines collectively called the CALVEC subprogram library. The purpose of the CALVEC software library is the emulation of CALCOMP pen and ink graphics on a DECsystem 10. A user level interface with CALVEC software allows standard CALCOMP subprogram calls to produce a VECtor file, FOR24.VEC. This vector file may subsequently be postprocessed into an image in a variety of ways.
  • Fast, digital signal processing (DSP) has many applications. Typical hardware options for performing DSP are field-programmable gate arrays (FPGAs), application-specific integrated DSP chips, or general purpose personal computer systems. This paper presents a novel DSP platform that has been developed for feedback control on the HBT-EP tokamak device. The system runs all signal processing exclusively on a Graphics Processing Unit (GPU) to achieve real-time performance with latencies below 8 μs. Signals are transferred into and out of the GPU using PCI Express peer-to-peer direct-memory-access transfers without involvement of the central processing unit or host memory. Tests were performed on themore » feedback control system of the HBT-EP tokamak using forty 16-bit floating point inputs and outputs each and a sampling rate of up to 250 kHz. Signals were digitized by a D-TACQ ACQ196 module, processing done on an NVIDIA GTX 580 GPU programmed in CUDA, and analog output was generated by D-TACQ AO32CPCI modules.« less

To initiate an order for this software, request consultation services, or receive further information, fill out the request form below. You may also reach us by email at: .

OSTI staff will begin to process an order for scientific and technical software once the payment and signed site license agreement are received. If the forms are not in order, OSTI will contact you. No further action will be taken until all required information and/or payment is received. Orders are usually processed within three to five business days.

Software Request

(required)
(required)
(required)
(required)
(required)
(required)
(required)
(required)