Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “message passing interface”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Machine Learning Methods for Connection RTT and Loss Rate Estimation Using MPI Measurements Under Random Losses

Scientific computations are expected to be increasingly distributed across wide-area networks, and Message Passing Interface (MPI) has been shown to scale to support their communications over long distances. Application-level measurements of MPI operations reflect the connection Round-Trip Time (RTT) and loss rate, and machine learningmethods have been previously developed to estimate them under deterministic periodic losses. In this paper, we consider more complex, random losses with uniform, Poisson and Gaussian distributions. We study five disparate machine leaning methods, with linear and non-linear, and smooth and non-smooth properties, to estimate RTT and loss rate over 10 Gbps connections with 0–366 ms RTT. The diversity and complexity of these estimators combined with the randomness of losses and TCP’s non-linear response together rule out the selection of a single best among them; instead, we fuse them to retain their design diversity. Overall, the results show that accurate estimates can be generated at low loss ratesbut become inaccurate at loss rates 10% and higher, thereby illustrating both their strengths and limitations.

Rao, Nageswara S.↗
DOE OSTI2020

TriC: Distributed-memory Triangle Counting by Exploiting the Graph Structure

Graph analytics has emerged as an important tool in the analysis of large scale data from diverse application domains such as social networks, cyber security and bioinformatics. Counting the number of triangles in a graph is a fundamental kernel with several applications such as detecting the community structure of a graph or in identifying important vertices in a graph. The ubiquity of massive datasets is driving the need to scale graph analytics on parallel systems. However, numerous challenges exist in efficiently parallelizing graph algorithms, especially on distributed-memory systems. Irregular memory accesses and communication patterns, low computation to communication ratios, and the need for frequent synchronization are some of the leading challenges. In this paper, we present TriC, our distributed-memory implementation of triangle counting in graphs using the Message Passing Interface (MPI), as a submission to the 2020 GraphChallenge competition. Using a set of synthetic and real-world inputs from the challenge, we demonstrate a speedup of up to 90x relative to previous work on 32 processor-cores of a NERSC Cori node. We also provide details from distributed runs with up to8192 processes along with strong scaling results. The observations presented in this work provide an understanding of the system-level bottlenecks at scale that specifically impact sparse-irregular workloads and will therefore benefit other efforts to parallelize graph algorithms.

Halappanavar, Mahantesh↗
DOE OSTI2023

Software Quality Assurance for High Performance Computing Containers

Software containers are a key channel for delivering portable and reproducible scientific software in high performance computing (HPC) environments. HPC environments are different from other types of computing environments primarily due to usage of the message passing interface (MPI) and drivers for specialized hard- ware to enable distributed computing capabilities. This distinction directly impacts how software containers are built for HPC applications and can complicate software quality assurance efforts including portability and performance. This work introduces a strategy for building containers for HPC applications that adopts layering as a mechanism for software quality assurance. The strategy is demonstrated across three different HPC systems, two of them petaflops scale with entirely different interconnect technologies and/or processor chipsets but running the same container. Performance consequences of the containerization strategy are found to be less than 5-14% while still achieving portable and reproducible containers for HPC systems.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1997

Cooperative Data Sharing: Simple Support for Clusters of SMP Nodes

Libraries like PVM and MPI send typed messages to allow for heterogeneous cluster computing. Lower-level libraries, such as GAM, provide more efficient access to communication by removing the need to copy messages between the interface and user space in some cases. still lower-level interfaces, such as UNET, get right down to the hardware level to provide maximum performance. However, these are all still interfaces for passing messages from one process to another, and have limited utility in a shared-memory environment, due primarily to the fact that message passing is just another term for copying. This drawback is made more pertinent by today's hybrid architectures (e.g. clusters of SMPs), where it is difficult to know beforehand whether two communicating processes will share memory. As a result, even portable language tools (like HPF compilers) must either map all interprocess communication, into message passing with the accompanying performance degradation in shared memory environments, or they must check each communication at run-time and implement the shared-memory case separately for efficiency. Cooperative Data Sharing (CDS) is a single user-level API which abstracts all communication between processes into the sharing and access coordination of memory regions, in a model which might be described as "distributed shared messages" or "large-grain distributed shared memory". As a result, the user programs to a simple latency-tolerant abstract communication specification which can be mapped efficiently to either a shared-memory or message-passing based run-time system, depending upon the available architecture. Unlike some distributed shared memory interfaces, the user still has complete control over the assignment of data to processors, the forwarding of data to its next likely destination, and the queuing of data until it is needed, so even the relatively high latency present in clusters can be accomodated. CDS does not require special use of an MMU, which can add overhead to some DSM systems, and does not require an SPMD programming model. unlike some message-passing interfaces, CDS allows the user to implement efficient demand-driven applications where processes must "fight" over data, and does not perform copying if processes share memory and do not attempt concurrent writes. CDS also supports heterogeneous computing, dynamic process creation, handlers, and a very simple thread-arbitration mechanism. Additional support for array subsections is currently being considered. The CDS1 API, which forms the kernel of CDS, is built primarily upon only 2 communication primitives, one process initiation primitive, and some data translation (and marshalling) routines, memory allocation routines, and priority control routines. The entire current collection of 28 routines provides enough functionality to implement most (or all) of MPI 1 and 2, which has a much larger interface consisting of hundreds of routines. still, the API is small enough to consider integrating into standard os interfaces for handling inter-process communication in a network-independent way. This approach would also help to solve many of the problems plaguing other higher-level standards such as MPI and PVM which must, in some cases, "play OS" to adequately address progress and process control issues. The CDS2 API, a higher level of interface roughly equivalent in functionality to MPI and to be built entirely upon CDS1, is still being designed. It is intended to add support for the equivalent of communicators, reduction and other collective operations, process topologies, additional support for process creation, and some automatic memory management. CDS2 will not exactly match MPI, because the copy-free semantics of communication from CDS1 will be supported. CDS2 application programs will be free to carefully also use CDS1. CDS1 has been implemented on networks of workstations running unmodified Unix-based operating systems, using UDP/IP and vendor-supplied high- performance locks. Although its inter-node performance is currently unimpressive due to rudimentary implementation technique, it even now outperforms highly-optimized MPI implementation on intra-node communication due to its support for non-copy communication. The similarity of the CDS1 architecture to that of other projects such as UNET and TRAP suggests that the inter-node performance can be increased significantly to surpass MPI or PVM, and it may be possible to migrate some of its functionality to communication controllers.

DiNucci, David C.↗
NASA NTRS1996

Parallel NPARC: Implementation and Performance

Version 3 of the NPARC Navier-Stokes code includes support for large-grain (block level) parallelism using explicit message passing between a heterogeneous collection of computers. This capability has the potential for significant performance gains, depending upon the block data distribution. The parallel implementation uses a master/worker arrangement of processes. The master process assigns blocks to workers, controls worker actions, and provides remote file access for the workers. The processes communicate via explicit message passing using an interface library which provides portability to a number of message passing libraries, such as PVM (Parallel Virtual Machine). A Bourne shell script is used to simplify the task of selecting hosts, starting processes, retrieving remote files, and terminating a computation. This script also provides a simple form of fault tolerance. An analysis of the computational performance of NPARC is presented, using data sets from an F/A-18 inlet study and a Rocket Based Combined Cycle Engine analysis. Parallel speedup and overall computational efficiency were obtained for various NPARC run parameters on a cluster of IBM RS6000 workstations. The data show that although NPARC performance compares favorably with the estimated potential parallelism, typical data sets used with previous versions of NPARC will often need to be reblocked for optimum parallel performance. In one of the cases studied, reblocking increased peak parallel speedup from 3.2 to 11.8.

Townsend, S. E.↗
DOE OSTI2024

Performance on HPC Platforms Is Possible Without C++

Computing at large scales has become extremely challenging due to increasing heterogeneity in both hardware and software. More and more scientific workflows must tackle a range of scales and use machine learning and AI intertwined with more traditional numerical modeling methods, placing more demands on computational platforms. These constraints indicate a need to fundamentally rethink the way computational science is done and the tools that are needed to enable these complex workflows. The current set of C++-based solutions may not suffice, and relying exclusively upon C++ may not be the best option, especially because several newer languages and boutique solutions offer more robust design features to tackle the challenges of heterogeneity. In June 2023, we held a mini symposium that explored the use of newer languages and heterogeneity solutions that are not tied to C++ and that offer options beyond template metaprogramming and Parallel. For for performance and portability. In conclusion, we describe some of the presentations and discussion from the mini symposium in this article.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Bringing HPE Slingshot 11 support to Open MPI

The Cray HPE Slingshot 11 network is used on the new exascale systems arriving at the U.S. Department of Energy (DoE) laboratories (e.g., Frontier, Aurora, Perlmutter). As such, the support of this network is an important capability to meet the needs of exascale applications. Here, this article highlights recent work to develop supporting infrastructure to enable Open MPI to efficiently support these new platforms. A key component of this effort involves development of a new Open Fabrics Interface (OFI) provider, LinkX. We discuss the design and development of enhancements that take advantage of the new Slingshot 11 network and AMD GPUs. We include performance data from tests on the Frontier supercomputer using synthetic communication benchmarks, and the vendor provided MPI as a baseline for comparison. The tests demonstrate full functionality of Open MPI on the system and initial results show favorable performance when compared to the highly tuned vendor implementation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Taking the MPI standard and the open MPI library to exascale

The Open MPI for Exascale (OMPI-X) project was one of two in the Exascale Computing Project (ECP) focused on advancing the MPI ecosystem. The OMPI-X team worked with other MPI Forum members to champion several important features for inclusion in the MPI 4.0, 4.1, and upcoming 5.0 MPI standard versions, in support of the needs of exascale applications and systems. The team also worked with the larger Open MPI community to bring implementations of these new features and other enhancements into Open MPI, one of the leading open-source implementations of the MPI interface. Here, this paper describes the motivation for the work of the OMPI-X project in the context of exascale computing needs, the nature of the resulting new capabilities in the MPI standard, and how they were implemented in the Open MPI library. Features include improved support for “MPI + X” programming models through partitioned communications and support for user-level threading, sessions, fault tolerance through the user-level fault mitigation (ULFM) and Reinit models, and other features. We also discuss enhancements to Open MPI providing improved performance and scalability for existing features, such as collective operations, one-sided operations, support for the Slingshot-11 interconnect of the initial exascale systems, and how the OMPI-X team worked to improve quality assurance for the Open MPI library, particularly on platforms of interest to the Department of Energy community.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Parallel Distributed Hydrology Soil Vegetation Model (DHSVM) using global arrays

The Distributed Hydrology Soil Vegetation Model (DHSVM) code was parallelized for distributed memory computers using Global Arrays (GA). To analyze parallel performance, DHSVM was used to simulate two significantly sized river basins located in northwest continental United States and southwest Canada at 90~m resolution: (1) Clearwater (25,000~km2) and (2) Columbia (668,000~km2). Meteorological forcing applied to both basins was dynamically down-scaled from a global circulation model using the Weather Research and Forecasting (WRF) model and read into DHSVM as 2D maps each time step. We report parallel code speed up was significant. Run times for 1-year simulations were reduced by an order of magnitude for both test basins. A maximum speed up of 105 was attained with 480 processors while simulating the Columbia basin. Speed up was limited by input dominated tasks, particularly the input of meteorological forcing data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

GPU acceleration of local and semilocal density functional calculations in the SPARC electronic structure code

We present a Graphics Processing Unit (GPU)-accelerated version of the real-space SPARC electronic structure code for performing Kohn–Sham density functional theory calculations within the local density and generalized gradient approximations. In particular, we develop a modular math-kernel based implementation for NVIDIA architectures wherein the computationally expensive operations are carried out on the GPUs, with the remainder of the workload retained on the central processing units (CPUs). Here, using representative bulk and slab examples, we show that relative to CPU-only execution, GPUs enable speedups of up to 6× and 60× in node and core hours, respectively, bringing time to solution down to less than 30 s for a metallic system with over 14 000 electrons and enabling significant reductions in computational resources required for a given wall time.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

GPU acceleration of hybrid functional calculations in the SPARC electronic structure code

We present a Graphics Processing Unit (GPU)-accelerated version of the real-space SPARC electronic structure code for performing hybrid functional calculations in generalized Kohn–Sham density functional theory. In particular, we develop a batch variant of the recently formulated Kronecker product-based linear solver for the simultaneous solution of multiple linear systems. We then develop a modular, math kernel based implementation for hybrid functionals on NVIDIA architectures, where computationally intensive operations are offloaded to the GPUs, while the remaining workload is handled by the central processing units (CPUs). Considering bulk and slab examples, we demonstrate that GPUs enable up to 8× speedup in node-hours and 80× in core-hours compared to CPU-only execution, reducing the time to solution on V100 GPUs to around 300 s for a metallic system with over 6000 electrons, and significantly reducing the computational resources required for a given wall time.

Kohn-Sham density functional theory↗
DOE OSTI2020

Generating Massive Scale-free Networks: Novel Parallel Algorithms using the Preferential Attachment Model

Recently, there has been substantial interest in the study of various random networks as mathematical models of complex systems. As real-life complex systems grow larger, the ability to generate progressively large random networks becomes all the more important. This motivates the need for efficient parallel algorithms for generating such networks. Naïve parallelization of sequential algorithms for generating random networks is inefficient due to inherent dependencies among the edges and the possibility of creating duplicate (parallel) edges. In this article, we present message passing interface-based distributed memory parallel algorithms for generating random scale-free networks using the preferential-attachment model. Our algorithms are experimentally verified to scale very well to a large number of processing elements (PEs), providing near-linear speedups. The algorithms have been exercised with regard to scale and speed to generate scale-free networks with one trillion edges in 6 minutes using 1,000 PEs.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2003

SpaceWire Driver Software for Special DSPs

A computer program provides a high-level C-language interface to electronics circuitry that controls a SpaceWire interface in a system based on a space qualified version of the ADSP-21020 digital signal processor (DSP). SpaceWire is a spacecraft-oriented standard for packet-switching data-communication networks that comprise nodes connected through bidirectional digital serial links that utilize low-voltage differential signaling (LVDS). The software is tailored to the SMCS-332 application-specific integrated circuit (ASIC) (also available as the TSS901E), which provides three highspeed (150 Mbps) serial point-to-point links compliant with the proposed Institute of Electrical and Electronics Engineers (IEEE) Standard 1355.2 and equivalent European Space Agency (ESA) Standard ECSS-E-50-12. In the specific application of this software, the SpaceWire ASIC was combined with the DSP processor, memory, and control logic in a Multi-Chip Module DSP (MCM-DSP). The software is a collection of low-level driver routines that provide a simple message-passing application programming interface (API) for software running on the DSP. Routines are provided for interrupt-driven access to the two styles of interface provided by the SMCS: (1) the "word at a time" conventional host interface (HOCI); and (2) a higher performance "dual port memory" style interface (COMI).

Clark, Douglas↗
NASA NTRS1991

The AI Bus architecture for distributed knowledge-based systems

The AI Bus architecture is layered, distributed object oriented framework developed to support the requirements of advanced technology programs for an order of magnitude improvement in software costs. The consequent need for highly autonomous computer systems, adaptable to new technology advances over a long lifespan, led to the design of an open architecture and toolbox for building large scale, robust, production quality systems. The AI Bus accommodates a mix of knowledge based and conventional components, running on heterogeneous, distributed real world and testbed environment. The concepts and design is described of the AI Bus architecture and its current implementation status as a Unix C++ library or reusable objects. Each high level semiautonomous agent process consists of a number of knowledge sources together with interagent communication mechanisms based on shared blackboards and message passing acquaintances. Standard interfaces and protocols are followed for combining and validating subsystems. Dynamic probes or demons provide an event driven means for providing active objects with shared access to resources, and each other, while not violating their security.

Schultz, Roger D.↗
NASA NTRS1995

MPI-IO: A Parallel File I/O Interface for MPI Version 0.3

Thanks to MPI [9], writing portable message passing parallel programs is almost a reality. One of the remaining problems is file I/0. Although parallel file systems support similar interfaces, the lack of a standard makes developing a truly portable program impossible. Further, the closest thing to a standard, the UNIX file interface, is ill-suited to parallel computing. Working together, IBM Research and NASA Ames have drafted MPI-I0, a proposal to address the portable parallel I/0 problem. In a nutshell, this proposal is based on the idea that I/0 can be modeled as message passing: writing to a file is like sending a message, and reading from a file is like receiving a message. MPI-IO intends to leverage the relatively wide acceptance of the MPI interface in order to create a similar I/0 interface. The above approach can be materialized in different ways. The current proposal represents the result of extensive discussions (and arguments), but is by no means finished. Many changes can be expected as additional participants join the effort to define an interface for portable I/0. This document is organized as follows. The remainder of this section includes a discussion of some issues that have shaped the style of the interface. Section 2 presents an overview of MPI-IO as it is currently defined. It specifies what the interface currently supports and states what would need to be added to the current proposal to make the interface more complete and robust. The next seven sections contain the interface definition itself. Section 3 presents definitions and conventions. Section 4 contains functions for file control, most notably open. Section 5 includes functions for independent I/O, both blocking and nonblocking. Section 6 includes functions for collective I/O, both blocking and nonblocking. Section 7 presents functions to support system-maintained file pointers, and shared file pointers. Section 8 presents constructors that can be used to define useful filetypes (the role of filetypes is explained in Section 2 below). Section 9 presents how the error handling mechanism of MPI is supported by the MPI-IO interface. All this is followed by a set of appendices, which contain information about issues that have not been totally resolved yet, and about design considerations. The reader can find there the motivation behind some of our design choices. More information on this would definitely be welcome and will be included in a further release of this document. The first appendix contains a description of MPI-I0's 'hints' structure which is used when opening a file. Appendix B is a discussion of various issues in the support for file pointers. Appendix C explains what we mean in talking about atomic access. Appendix D provides detailed examples of filetype constructors, and Appendix E contains a collection of arguments for and against various design decisions.

Corbett, Peter↗
NASA NTRS1996

Performance Evaluation of Communication Software Systems for Distributed Computing

In recent years there has been an increasing interest in object-oriented distributed computing since it is better quipped to deal with complex systems while providing extensibility, maintainability, and reusability. At the same time, several new high-speed network technologies have emerged for local and wide area networks. However, the performance of networking software is not improving as fast as the networking hardware and the workstation microprocessors. This paper gives an overview and evaluates the performance of the Common Object Request Broker Architecture (CORBA) standard in a distributed computing environment at NASA Ames Research Center. The environment consists of two testbeds of SGI workstations connected by four networks: Ethernet, FDDI, HiPPI, and ATM. The performance results for three communication software systems are presented, analyzed and compared. These systems are: BSD socket programming interface, IONA's Orbix, an implementation of the CORBA specification, and the PVM message passing library. The results show that high-level communication interfaces, such as CORBA and PVM, can achieve reasonable performance under certain conditions.

Fatoohi, Rod↗
NASA NTRSDate not supplied

Software for Optical (Laser) Ground Station Monitor and Control ​

Previous NASA laser communication missions have been supported by ground terminals specific to the mission. The Low-Cost Optical Terminal project (LCOT) aims to serve as a commercial off-the-shelf (COTS), reusable, and modular optical ground terminal prototype, provide a blueprint for future optical ground terminals, and enable optical communication experiments with a variety of spacecraft from Low Earth Orbit to lunar orbit. The goal of the internship was the development of LCOT’s Gimbal Monitor and Control (GMC) application, within the LCOT Monitor and Control Subsystem (MCS). Mount control software PWI4 was provided by mount and gimbal vendor Planewave Instruments; developed in Python, GMC integrates and interfaces with PWI4 using third party libraries such as Protobuf and RabbitMQ. As a stand in for the Monitor and Control Subsystem (MCS), a test Graphical User Interface (GUI) was created to send commands to and receive telemetry from the GMC application; these commands and telemetry are sent through the RabbitMQ message bus as Protobuf encoded messages. GMC then interfaces with PWI4 which passes along desired commands and telemetry to and from a vendor provided mount and gimbal simulator. The GMC software developed allows LCOT’s Monitor and Control Subsystem (MCS) to take advantage of the existing mount control software, advancing LCOT’s efforts in the development of the MCS. The MCS and GMC software developed will contribute to LCOT’s goal as a flexible and modular optical ground terminal prototype and blueprint, which supports development towards a potential optical ground terminal network.

space communications↗