Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel systems”

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 325 records · Page 18

Large Scale Caching and Streaming of Training Data for Online Deep Learning

The training of deep neural network models on large data remains a difficult problem, despite progress towards scalable techniques. In particular, there is a mismatch between the random but predetermined order in which AI flows select training samples and the streaming I/O patterns for which traditional HPC data storage (e.g., parallel file systems) are designed. In addition, as more data are obtained, it is feasible neither simply to train learning models incrementally, due to catastrophic forgetting (i.e., bias towards new samples), nor to train frequently from scratch, due to prohibitive time and/or resource constraints. In this paper, we study data management techniques that combine caching and streaming with rehearsal support in order to enable efficient access to training samples in both offline training and continual learning. We revisit state-of-art streaming approaches based on data pipelines that transparently handle prefetching, caching, shuffling, and data augmentation, and discuss the challenges and opportunities that arise when combining these methods with data-parallel training techniques. We also report on preliminary experiments that evaluate the I/O overheads involved in accessing the training samples from a parallel file system (PFS) under several concurrency scenarios, highlighting the impact of the PFS on the design of the data pipelines.

data pipelines↗

Redundant Execution of HPC Applications with MR-MPI

This paper presents a modular-redundant Message Passing Interface (MPI) solution, MR-MPI, for transparently executing high-performance computing (HPC) applications in a redundant fashion. The presented work addresses the deficiencies of recovery-oriented HPC, i.e., checkpoint/restart to/from a parallel file system, at extreme scale by adding the redundancy approach to the HPC resilience portfolio. It utilizes the MPI performance tool interface, PMPI, to transparently intercept MPI calls from an application and to hide all redundancy-related mechanisms. A redundantly executed application runs with r*m native MPI processes, where r is the number of MPI ranks visible to the application and m is the replication degree. Messages between redundant nodes are replicated. Partial replication for tunable resilience is supported. The performance results clearly show the negative impact of the O(m^m) messages between replicas. For low-level, point-to-point benchmarks, the impact can be as high as the replication degree. For applications, performance highly depends on the actual communication types and counts. On single-core systems, the overhead can be 0% for embarrassingly parallel applications independent of the employed redundancy configuration or up to 70-90% for communication-intensive applications in a dual-redundant configuration. On multi-core systems, the overhead can be significantly higher due to the additional communication contention.

Engelmann, Christian↗

The OVIS analysis architecture.

This report summarizes the current statistical analysis capability of OVIS and how it works in conjunction with the OVIS data readers and interpolators. It also documents how to extend these capabilities. OVIS is a tool for parallel statistical analysis of sensor data to improve system reliability. Parallelism is achieved using a distributed data model: many sensors on similar components (metaphorically sheep) insert measurements into a series of databases on computers reserved for analyzing the measurements (metaphorically shepherds). Each shepherd node then processes the sheep data stored locally and the results are aggregated across all shepherds. OVIS uses the Visualization Tool Kit (VTK) statistics algorithm class hierarchy to perform analysis of each process's data but avoids VTK's model aggregation stage which uses the Message Passing Interface (MPI); this is because if a single process in an MPI job fails, the entire job will fail. Instead, OVIS uses asynchronous database replication to aggregate statistical models. OVIS has several additional features beyond those present in VTK that, first, accommodate its particular data format and, second, improve the memory and speed of the statistical analyses. First, because many statistical algorithms are multivariate in nature and sensor data is typically univariate, interpolation of data is required to provide simultaneous observations of metrics. Note that in this report, we will refer to a single value obtained from a sensor as a measurement while a collection of multiple sensor values simultaneously present in the system is an observation. A base class for interpolation is provided that abstracts the operation of converting multiple sensor measurements into simultaneous observations. A concrete implementation is provided that performs piecewise constant temporal interpolation of multiple metrics across a single component. Secondly, because calculations may summarize data too large to fit in memory OVIS analyses batches of observations at a time and aggregates these intermediate intra-process models as it goes before storing the final model for inter-process aggregation via database replication. This reduces the memory footprint of the analysis, interpolation, and the database client and server query processing. This also interleaves processing with the disk I/O required to fetch data from the database - also improving speed. This report documents how OVIS performs analyses and how to create additional analysis components that fetch measurements from the database, perform interpolation, or perform operations on streamed observations (such as model updates or assessments). The rest of this section outlines the OVIS analysis algorithm and is followed by sections specific to each subtask. Note that we are limiting our discussion for now to the creation of a model from a set of measurements, and not including the assessment of observations using a model. The same framework can be used for assessment but that use case is not detailed in this report.

97 MATHEMATICS AND COMPUTING↗

Storage of sparse files using parallel log-structured file system

A sparse file is stored without holes by storing a data portion of the sparse file using a parallel log-structured file system; and generating an index entry for the data portion, the index entry comprising a logical offset, physical offset and length of the data portion. The holes can be restored to the sparse file upon a reading of the sparse file. The data portion can be stored at a logical end of the sparse file. Additional storage efficiency can optionally be achieved by (i) detecting a write pattern for a plurality of the data portions and generating a single patterned index entry for the plurality of the patterned data portions; and/or (ii) storing the patterned index entries for a plurality of the sparse files in a single directory, wherein each entry in the single directory comprises an identifier of a corresponding sparse file.

97 MATHEMATICS AND COMPUTING↗

Plasmon hybridization in parallel nano-wire systems

We apply the plasmon hybridization method to a double-nano-wire system, providing a simple and intuitive description of the plasmon excitations in the system. We apply the two-center cylindrical coordinate system for mathematical convenience and find an explicit form of the surface plasmon oscillations, in terms of the interaction between the bare plasmon modes of the individual surfaces of the nano-wires. We present numerical results to display how the plasmon excitations of the system depend on nano-wire separation when there is no angular momentum transfer, i.e., when m = 0.

36 MATERIALS SCIENCE↗

Containers for Massive Ensemble of I/O Bound Hierarchical Coupled Simulations

We present our experience using containers to scale up a massive ensemble of coupled I/O bound workloads on the NERSC Cori supercomputer. We describe the design of a hierarchical simulation structure using the Integrated Plasma Simulator (IPS) that enables the flexible execution of coupled simulations at the system, node, and core level using the same coupling abstraction and API. The hierarchical design allows for the node-level execution to be efficiently executed using containers while not impacting the structure of the simulation at the system level. We demonstrate the viability of the approach by presenting experimental results from applications in coupled fusion plasma simulations that illustrate the performance impact of using containers to deploy the node-level workloads, in conjunction with the user mountable XFS file systems to ameliorate the load on the Lustre parallel file system. We also present results from production runs showing the ability of the ensemble simulations to scale to hundreds of Cori Haswell nodes, with little or no overhead.

Elwasif, Wael↗

The Portals 4.1 Network Programming Interface

This report presents a specification for the Portals 4 networ k programming interface. Portals 4 is intended to allow scalable, high-performance network communication betwee n nodes of a parallel computing system. Portals 4 is well suited to massively parallel processing and embedded syste ms. Portals 4 represents an adaption of the data movement layer developed for massively parallel processing platfor ms, such as the 4500-node Intel TeraFLOPS machine. Sandia's Cplant cluster project motivated the development of Version 3.0, which was later extended to Version 3.3 as part of the Cray Red Storm machine and XT line. Version 4 is tar geted to the next generation of machines employing advanced network interface architectures that support enh anced offload capabilities.

97 MATHEMATICS AND COMPUTING↗

Incorporating Human Readiness Levels at Sandia National Laboratories

Since 2010, the concept of human readiness levels has been under development as a possible supplement to the existing technology readiness level (TRL) scale. The intent is to provide a mechanism to address safety and performance risks associated with the human component in a system that parallels the TRL structure already familiar to the systems engineering community. Sandia National Laboratories in Albuquerque, New Mexico, initiated a study in 2015 to evaluate options to incorporate human readiness planning for Sandia processes and products. The study team has collected the majority of baseline assessment data and has conducted interviews to understand staff perceptions of four different options for human readiness planning. Preliminary results suggest that all four options may have a vital role, depending on the type of work performed and the phase of product development. Upon completion of data collection, the utility of identified solutions will be assessed in one or more test cases.

99 GENERAL AND MISCELLANEOUS↗

Ex-vessel neutron dosimetry analysis for westinghouse 4-loop XL pressurized water reactor plant using the RadTrack{sup TM} Code System with the 3D parallel discrete ordinates code RAPTOR-M3G

Traditional two-dimensional (2D)/one-dimensional (1D) SYNTHESIS methodology has been widely used to calculate fast neutron (>1.0 MeV) fluence exposure to reactor pressure vessel in the belt-line region. However, it is expected that this methodology cannot provide accurate fast neutron fluence calculation at elevations far above or below the active core region. A three-dimensional (3D) parallel discrete ordinates calculation for ex-vessel neutron dosimetry on a Westinghouse 4-Loop XL Pressurized Water Reactor has been done. It shows good agreement between the calculated results and measured results. Furthermore, the results show very different fast neutron flux values at some of the former plate locations and elevations above and below an active core than those calculated by a 2D/1D SYNTHESIS method. This indicates that for certain irregular reactor internal structures, where the fast neutron flux has a very strong local effect, it is required to use a 3D transport method to calculate accurate fast neutron exposure. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

System and method for parallel two-photon lithography using a metalens array

A metalens array is disclosed for controllably modifying a phase of a wavefront of an optical beam. The metalens array may have a substrate having at least first and second metalens unit cells, and forming a single integrated structure with no stitching being required of the first and second metalens unit cells. The first metalens unit cell has a first plurality of nanoscale features and is configured to modify a phase of a first portion of a wavefront of an optical signal incident thereon in accordance with a first predetermined phase pattern to create at least one first focal voxel within an image plane. The second metalens unit cell has a second plurality of nanoscale features configured to modify the phase of a second portion of the wavefront of the optical signal incident thereon, in accordance with a second predetermined phase pattern, to simultaneously create at least one second focal voxel within the image plane. Each metalens unit cell also has an overall diameter of no more than about 200 microns.

Xia, Xiaoxing↗

An extensible operating system design for large-scale parallel machines.

Running untrusted user-level code inside an operating system kernel has been studied in the 1990's but has not really caught on. We believe the time has come to resurrect kernel extensions for operating systems that run on highly-parallel clusters and supercomputers. The reason is that the usage model for these machines differs significantly from a desktop machine or a server. In addition, vendors are starting to add features, such as floating-point accelerators, multicore processors, and reconfigurable compute elements. An operating system for such machines must be adaptable to the requirements of specific applications and provide abstractions to access next-generation hardware features, without sacrificing performance or scalability.

43 PARTICLE ACCELERATORS↗

An application-level parallel I/O library for Earth system models

We describe the design and implementation of an application-level parallel I/O (PIO) library for the reading and writing of distributed arrays to several common scientific data formats. PIO provides the flexibility to control the number of I/O tasks through data rearrangement to an I/O friendly decomposition. This flexibility enables reductions in per task memory usage and improvements in disk I/O performance versus a serial I/O approach. We illustrate the impact various features within PIO have on memory usage and disk I/O bandwidth on a Cray XT5 system.

Computer Science↗