Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “RPC”

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

UPC++ v1.0 Specification (Revision 2019.9.0)

UPC++ is a C++11 library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. We are revising the library under the auspices of the DOE’s Exascale Computing Project, to meet the needs of applications requiring PGAS support. UPC++ is intended for implementing elaborate distributed data structures where communication is irregular or fine-grained. The UPC++ interfaces for moving non-contiguous data and handling memories with different optimal access methods are composable and similar to those used in conventional C++. The UPC++ programmer can expect communication to run at close to hardware speeds. The key facilities in UPC++ are global pointers, that enable the programmer to express ownership information for improving locality, one-sided communication, both put/get and RPC, futures and continuations. Futures capture data readiness state, which is useful in making scheduling decisions, and continuations provide for completion handling via callbacks. Together, these enable the programmer to chain together a DAG of operations to execute asynchronously as high-latency dependencies become satisfied.

97 MATHEMATICS AND COMPUTING↗

An Update to the Letter of Intent for MATHUSLA: Search for Long-Lived Particles at the HL-LHC

We report on recent progress in the design of the proposed MATHUSLA Long Lived Particle (LLP) detector for the HL-LHC, updating the information in the original Letter of Intent (LoI), see CDS:LHCC-I-031, arXiv:1811.00927. A suitable site has been identified at LHC Point 5 that is closer to the CMS Interaction Point (IP) than assumed in the LoI. The decay volume has been increased from 20 m to 25 m in height. Engineering studies have been made in order to locate much of the decay volume below ground, bringing the detector even closer to the IP. With these changes, a 100 m x 100 m detector has the same physics reach for large c$\tau$ as the 200 m x 200 m detector described in the LoI and other studies. The performance for small c$\tau$ is improved because of the proximity to the IP. Detector technology has also evolved while retaining the strip-like sensor geometry in Resistive Plate Chambers (RPC) described in the LoI. The present design uses extruded scintillator bars read out using wavelength shifting fibers and silicon photomultipliers (SiPM). Operations will be simpler and more robust with much lower operating voltages and without the use of greenhouse gases. Manufacturing is straightforward and should result in cost savings. Understanding of backgrounds has also significantly advanced, thanks to new simulation studies and measurements taken at the MATHUSLA test stand operating above ATLAS in 2018. We discuss next steps for the MATHUSLA collaboration, and identify areas where new members can make particularly important contributions.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

UPC++ v1.0 Specification, Revision 2020.10.0

UPC++ is a C++11 library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

An updated estimate of the Mu2e experiment sensitivity

The Mu2e experiment at Fermilab will search for the conversion of a negative muon into an electron inside the field of a nucleus. This process does not conserve charged-lepton flavour and is heavily suppressed in the Standard Model (SM), with a branching ratio < 10-50. Any evidence of it would be an undeniable evidence of new physics beyond the SM. The project sets out to achieve a single event sensitivity of $\sim 3 × 10^{-17}$ on the ratio between the probability for a conversion of a negative muon into an electron and the one for a muon capture by the nucleus. Such a sensitivity would represent a 4 orders of magnitude improvement on the previous upper limit for the process, making possible to test predictions of different extensions of the SM. Mu2e uses three superconducting solenoids to produce and measure the muon conversions. In the first solenoid, the Production Solenoid, pions and kaons are produced, together with other secondary products, by 8 GeV kinetic energy proton interactions in a tungsten target. A gradient magnetic field is specifically designed to direct low momentum particles into the Transport Solenoid, an S-shaped magnet that filters out particles with unwanted charge and momentum. Muons, produced by pion and kaon decays, finally reach the aluminum Stopping Target in the Detector Solenoid, where they eventually stop and convert. The result of the conversion process is a monochromatic electron of ~105 MeV/c momentum. The Detector Solenoid also hosts the two main detectors: a straw tube tracker and two CsI calorimeter disks, both providing measurement of event kinematics. A germanium detector and a LaBr crystal are located downstream of the Detector Solenoid to measure the X and gamma rays produced by the muon captures in the Stopping Target. A veto system of scintillators covering the Detector Solenoid and half of the Transport Solenoid is used to identify and reject cosmic rays interactions. With respect to the initial project, the Mu2e running plan has evolved to a staged configuration with 2 years at reduced intensity before the 2025 accelerator shutdown for the neutrino beam upgrade and 2 or 3 years at full intensity after that. This, together with geometry changes and a better knowledge of detector performances obtained by the first slice tests, has required a full revision of the signal over background selection that is the subject of this thesis. In order to achieve a new estimate of Mu2e sensitivity, the simulation of the data corresponding to the first 2 years of data acquisition has been performed. This includes both conversion electrons (CE) and the main sources of background: cosmics, decay-in orbits (DIO), radiative pion captures (RPC) and antiprotons. The characteristics of the signal and of the main backgrounds have been studied to define the best selection variables for CE. A special effort has been devoted to the evaluation of the antiproton background. The lack of experimental data for antiproton production cross section makes the systematic uncertainty on this background significant. A new parameterization of the cross section has been developed to fit the existing data and to provide a more reliable estimate of the systematic uncertainty by comparing the results of the old and the new model. A special effort has been devoted to the optimization of the antiprotons Monte Carlo generator. This study has also revealed that the dominant component within this background is represented by antiprotons produced in the opposite direction with respect to the Transport Solenoid entrance and then redirected to it by back-scattering processes in the Production Target. This ultimately highlights the sensitivity of the background estimate to the G4 handling of antiproton interactions in the 1 to 3 GeV/c momentum range. Finally, the final experimental sensitivity has been studied. The momentum and time selection have been optimized to obtain the 5-sigma discovery reach or, in case of no signal, the upper limit on conversion probability. The results confirm that, in the firs t two years of data taking, Mu2e will be able to improve the current experimental sensitivity for muon-to-electron conversion in an atomic field by more than 3 order of magnitudes.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

UPC++ v1.0 Programmer’s Guide, Revision 2021.9.0

UPC++ is a C++ library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. PGAS additionally provides one-sided Remote Memory Access (RMA) to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. In UPC++, all communication operations are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all communication operations are asynchronous by default, to enable programmers to write code that scales well even on hundreds of thousands of cores.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

UPC++ v1.0 Specification, Revision 2021.9.0

UPC++ is a C++ library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Distributed Training for High Resolution Images: A Domain and Spatial Decomposition Approach

In this work we developed two Pytorch libraries using the PyTorch RPC interface for distributed deep learning approaches on high resolution images. The spatial decomposition library allows for distributedtraining on very large images, which otherwise won’t be possible on a single GPU. The domain parallelism library allows for distributed training across multiple domain unlabeled data, by leveraging the domain separation architecture. Both of those libraries were tested on the Summit supercomputer at a moderate scale, and we are releasing the code for both of them.

97 MATHEMATICS AND COMPUTING↗

Comparative Study for Pressure Gain Combustion-Gas Turbine System Performance in NGCC Configurations

The Department of Energy (DOE) has a goal to increase natural gas combined cycle (NGCC) plant efficiency to greater than 70 percent on a lower heating value (LHV) basis. Improvement in gas turbine (GT) efficiency is key achieving this goal. While many of the traditional improvement methods (higher firing temperature, advanced materials, improved cooling methods, etc.) may increase current GT and NGCC efficiencies incrementally, combining these traditional advancements with breakthrough technologies such as pressure gain combustion (PGC) may provide significant system efficiency gains. PGC is a combustion process that results in a net pressure increase across the combustor (up to pressure ratio of 2.2) rather than the traditional (<5%) pressure drop. There are multiple PGC technologies being developed that create the pressure rise phenomena, including resonant pulse combustors (RPCs), pulse detonation engines (PDEs), wave rotor combustor (WRC) engines and rotating detonation engines (RDEs). In this study, scoping analyses of PGC concepts applied to NGCC power plants were performed by combining simplified analytical models for the RPC, PDE, WRC, and RDE types of pressure gain processes with a GT model to evaluate impacts of using each of the PGC technologies on key GT performance parameters. Sensitivity analyses were also performed. Ultimately, the PGC technologies were found to have the potential to increase simple cycle efficiency by 4–6 percent (absolute) and NGCC efficiency by 2–3 percent (absolute) based on the results of the current study.

03 NATURAL GAS↗

UPC++ v1.0 Specification (Rev. 2023.9.0)

UPC++ is a C++ library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

Report of the IMFIT Advisory Committee from the Meeting of 9/16/08 at General Atomics

The Committee commends and thanks the presenters and GA for providing clear and informative presentations. We present our responses to 5 questions raised in the Charge, as well as additional comments that may be useful for the project. (Question 1) The IMFIT framework is based on PYTHON and CCA; can IMFIT meet its long-term goals with this choice of language for framework? Our first recommendation is to change the terminology; it appears that the project is not really utilizing CCA (Common Component Architecture). The project is really about using Python to organize workflow among components that are spawned as processes via pipes from Python. We suggest that in describing the project, the team drop the designation 'CCA' and simply state that the IMFIT framework is based on using Python to control workflow among components. Concerning the specific question: we understand the principal goal of the project to be the development of an easy-to-use tool that efficiently integrates different physics modules to support experimental data analysis and modeling. Yes, Python and the workflow-based framework are appropriate choices for meeting this objective. The Committee, in examining the goal, discussed whether the IMFIT project was duplicative of other projects; there are several others in the U.S. and elsewhere with integrated modeling goals. In particular in the U.S., SWIM, PTRANSP, and the proto-FSP's. However this project occupies a reasonably unique niche through its emphases on integrating experimental data, use of existing components, its near-term focus, and the extent to which is targeted at experimentalist users. (Question 2) Will IMFIT benefit from the additional use of an interoperability language for framework such as BABEL? No, IMFIT is not really doing inter-language communication. In particular this will be the case as long as inter-component communication is via files--and that's OK so long as you aren't doing lots of short steps with communication at each step. (Question 3) IMFIT GUI is based on the public PyGTK toolkit; will IMFIT benefit from the additional use of other public Python graphic toolkits such as wxPython that is a cross-platform wrapper, or PyQt, or a commercial package such as IDL? PyGTK seems adequate for now. The team needs to decide if it needs more capability. Avoid licensed software that interferes with portability and the open-source goal (and thus in particular avoid IDL). (Question 4) Will IMFIT benefit from using a XML-RPC as alternative to sending files thru sockets? IMFIT could benefit by adopting some kind of self-describing format, such as XML or HDF5, but data can still be sent over sockets. This is part of a broader question: 'Should IMFIT adopt some kind of standard for communicating data between components?' To do anything other than what is done now requires modification of individual components (or convertors/wrappers). But moving to a self-describing standard is a good idea, which should be considered. You don't need to impose a standard, but you could decide on one and move toward it gradually, as the architecture doesn't impose a standard. But whether you adopt a standard and the pace of moving toward it is up to you. (Question 5) Will IMFIT benefit from any other available computational tool? (a) IMFIT would benefit from fuller use of an already incorporated tool--python language in the task execution specification, to enable desired features like branching; (b) If there is a high degree of data hierarchy, IMFIT should explore HDF5 as well as NETCDF and compare; (c) IMFIT would benefit from moving toward common data standards; (d) It would be useful for the IMFIT team to monitor developments in related projects such as the proto-FSP's and PTRANSP to ascertain if there are potentially useful tools.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Power Transfer Potential to the Southeast in Response to a Renewable Portfolio Standard: Final Report

Electricity consumption in the Southeastern US, including Florida, is approximately 32% of the total US. The availability of renewable resources for electricity production is relatively small compared to the high consumption. Therefore meeting a national renewable portfolio standard (RPS) is particularly challenging in this region. Neighboring regions, particularly to the west, have significant wind resources and given sufficient transmission these resources could serve energy markets in the SE. This report looks at renewable resource supply relative to demands and the potential for power transfer into the SE. We found that significant wind energy transfers, at the level of 30-60 GW, are expected to be economic in case of federal RPC or CO2 policy. Development of wind resources will depend not only on the available transmission capacity and required balancing resources, but also on electricity supply and demand factors.

24 POWER TRANSMISSION AND DISTRIBUTION↗

UPC++ v1.0 Specification (Revision 2022.3.0)

UPC++ is a C++ library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

UPC++ Specification v1.0, Draft 8

UPC++ is a C++11 library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. We are revising the library under the auspices of the DOE’s Exascale Computing Project, to meet the needs of applications requiring PGAS support. UPC++ is intended for implementing elaborate distributed data structures where communication is irregular or fine-grained. The UPC++ interfaces for moving non-contiguous data and handling memories with different optimal access methods are composable and similar to those used in conventional C++. The UPC++ programmer can expect communication to run at close to hardware speeds. The key facilities in UPC++ are global pointers, that enable the programmer to express ownership information for improving locality, one-sided communication, both put/get and RPC, futures and continuations. Futures capture data readiness state, which is useful in making scheduling decisions, and continuations provide for completion handling via callbacks. Together, these enable the programmer to chain together a DAG of operations to execute asynchronously as high-latency dependencies become satisfied.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Specification, Revision 2023.3.0

UPC++ is a C++ library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Specification (Revision 2021.3.0)

UPC++ is a C++11 library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Specification, Revision 2022.9.0

UPC++ is a C++ library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗

GASNet-EX Memory Kinds: Support for Device Memory in PGAS Programming Models

There is an emerging need for adaptive, lightweight communication in irregular HPC applications at exascale, where GPU accelerators provide the majority of available compute cycles. To address this need, Lawrence Berkeley National Lab is developing a programming system to support distributed-memory HPC application development using the Partitioned Global Address Space (PGAS) model. This work includes two major components: UPC++ and GASNet-EX. UPC++ is a C++ template library providing Remote Memory Access (RMA) and Remote Procedure Call (RPC) communication interfaces. GASNet-EX is a portable, high-performance communication middleware library, used by the implementations of UPC++ and many other PGAS programming models. We describe recent advances in GASNet-EX to efficiently implement zero-copy Remote Memory Access (RMA) communication to and from memory on accelerator devices such as GPUs. We demonstrate performance improvements via benchmark results from UPC++ (on Summit) and the Legion programming system (on DGX-1), both using GASNet-EX for communication.

Hargrove, Paul H↗

UPC++ v1.0 Specification, Revision 2020.3.0

UPC++ is a C++11 library providing classes and functions that support Partitioned Global Address Space (PGAS) programming. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). All communication operations are syntactically explicit and default to non-blocking; asynchrony is managed through the use of futures, promises and continuation callbacks, enabling the programmer to construct a graph of operations to execute asynchronously as high-latency dependencies are satisfied. A global pointer abstraction provides system-wide addressability of shared memory, including host and accelerator memories. The parallelism model is primarily process-based, but the interface is thread-safe and designed to allow efficient and expressive use in multi-threaded applications. The interface is designed for extreme scalability throughout, and deliberately avoids design features that could inhibit scalability.

97 MATHEMATICS AND COMPUTING↗