Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory 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 127 records · Page 7

Myrmics Memory Allocator

MMA is a stand-alone memory management system for MPI clusters. It implements a shared Partitioned Global Address Space, where multiple MPI processes request objects from the allocator and the latter provides them with system-wide unique memory addresses for each object. It provides applications with an intuitive way of managing the memory system in a unified way, thus enabling easier writing of irregular application code.

Lymperis, S.↗

Myrmics Memory Allocator

MMA is a stand-alone memory management system for MPI clusters. It implements a shared Partitioned Global Address Space, where multiple MPI processes request objects from the allocator and the latter provides them with system-wide unique memory addresses for each object. It provides applications with an intuitive way of managing the memory system in a unified way, thus enabling easier writing of irregular application code.

Lymperis, S.↗

A communication-avoiding 3D algorithm for sparse LU factorization on heterogeneous systems

We propose a new algorithm to improve the strong scalability of right-looking sparse LU factorization on distributed memory systems. Our 3D algorithm for sparse LU uses a three-dimensional MPI process grid, exploits elimination tree parallelism, and trades off increased memory for reduced per-process communication. We also analyze the asymptotic improvements for planar graphs (e.g., those arising from 2D grid or mesh discretizations) and certain non-planar graphs (specifically for 3D grids and meshes). For a planar graph with $n$ vertices, our algorithm reduces communication volume asymptotically in $n$ by a factor of $\mathscr{O}$ $\Big(\sqrt{log \ n}\Big)$ and latency by a factor of $\mathscr{O}$ $(log \ n)$. For nonplanar cases, our algorithm can reduce the per-process communication volume by 3× and latency by $\mathscr{O}$ $\Big(n^\frac13\Big)$ times. In all cases, the memory needed to achieve these gains is a constant factor. We implemented our algorithm by extending the 2D data structure used in SuperLU_DIST. Our new 3D code achieves empirical speedups up to 27× for planar graphs and up to 3.3× for non-planar graphs over the baseline 2D SuperLU_DIST when run on 24,000 cores of a Cray XC30. We extend the 3D algorithm for heterogeneous architectures by adding the Highly Asynchronous Lazy Offload (Halo) algorithm for co-processor offload. On 4096 nodes of a Cray XK7 with 32,768 CPU cores and 4096 Nvidia K20x GPUs, the 3D algorithm achieves empirical speedups up to 24× for planar graphs and 3.5× for non-planar graphs over the baseline 2D SuperLU_DIST with co-processor acceleration.

97 MATHEMATICS AND COMPUTING↗

Final Report for CCS Cross-Layer Reliability Visioning Study

The geometric rate of improvement of transistor size and integrated circuit performance known as Moore's Law has been an engine of growth for our economy, enabling new products and services, creating new value and wealth, increasing safety, and removing menial tasks from our daily lives. Affordable, highly integrated components have enabled both life-saving technologies and rich entertainment applications. Anti-lock brakes, insulin monitors, and GPS-enabled emergency response systems save lives. Cell phones, internet appliances, virtual worlds, realistic video games, and mp3 players enrich our lives and connect us together. Over the past 40 years of silicon scaling, the increasing capabilities of inexpensive computation have transformed our society through automation and ubiquitous communications. Looking forward, increasing unpredictability threatens our ability to continue scaling integrated circuits at Moore's Law rates. As the transistors and wires that make up integrated circuits become smaller, they display both greater differences in behavior among devices designed to be identical and greater vulnerability to transient and permanent faults. Conventional design techniques expend energy to tolerate this unpredictability by adding safety margins to a circuit's operating voltage, clock frequency or charge stored per bit. However, the rising energy costs needed to compensate for increasing unpredictability are rapidly becoming unacceptable in today's environment where power consumption is often the limiting factor on integrated circuit performance and energy efficiency is a national concern. Reliability and energy consumption are both reaching key inflection points that, together, threaten to reduce or end the benefits of feature size reduction. To continue beneficial scaling, we must use a cross-layer, Jull-system-design approach to reliability. Unlike current systems, which charge every device a substantial energy tax in order to guarantee correct operation in spite of rare events, such as one high-threshold transistor in a billion or one erroneous gate evaluation in an hour of computation, cross-layer reliability schemes make reliability management a cooperative effort across the system stack, sharing information across layers so that they only expend energy on reliability when an error actually occurs. Figure 1 illustrates an example of such a system that uses a combination of information from the application and cheap architecture-level techniques to detect errors. When an error occurs, mechanisms at higher levels in the stack correct the error, efficiently delivering correct operation to the user in spite of errors at the device or circuit levels. In the realms of memory and communication, engineers have a long history of success in tolerating unpredictable effects such as fabrication variability, transient upsets, and lifetime wear using information sharing, limited redundancy, and cross-layer approaches that anticipate, accommodate, and suppress errors. Networks use a combination of hardware and software to guarantee end-toend correctness. Error-detection and correction codes use additional information to correct the most common errors, single-bit transmission errors. When errors occur that cannot be corrected by these codes, the network protocol requests re-transmission of one or more packets until the correct data is received. Similarly, computer memory systems exploit a cross-layer division of labor to achieve high performance with modest hardware. Rather than demanding that hardware alone provide the virtual memory abstraction, software page-fault and TLB-miss handlers allow a modest piece of hardware, the TLB, to handle the common-case operations on a cyc1e-by-cycle basis while infrequent misses are handled in system software. Unfortunately, mitigating logic errors is not as simple or as well researched as memory or communication systems. This lack of understanding has led to very expensive solutions. For example, triple-modular redundancy masks errors by triplicating computations in either time or area. This mitigation methods imposes a 200% increase in energy consumption for every operation, not just the uncommon failure cases. At a time when computation is rapidly becoming part of our critical civilian and military infrastructure and decreasing costsfor computation are fueling our economy and our well being, we cannot afford increasingly unreliable electronics or a stagnation in capabilities per dollar, watt, or cubic meter. If researchers are able to develop techniques that tolerate the growing unpredictability of silicon devices, Moore's Law scaling should continue until at least 2022. During this 12-year time period, transistors, which are the building blocks of electronic devices, will scale their dimensions (feature sizes) from 45nm to 4.5nm.

99 GENERAL AND MISCELLANEOUS↗

Tag-Split Cache for Efficient GPGPU Cache Utilization

Modern GPUs employ cache to improve memory system efficiency. However, large amount of cache space is underutilized due to irregular memory accesses and poor spatial locality which exhibited commonly in GPU applications. Our experiments show that using smaller cache lines could improve cache space utilization, but it also frequently suffers from significant performance loss by introducing large amount of extra cache requests. In this work, we propose a novel cache design named tag-split cache (TSC) that enables fine-grained cache storage to address the problem of cache space underutilization while keeping memory request number unchanged. TSC divides tag into two parts to reduce storage overhead, and it supports multiple cache line replacement in one cycle.

performance and energy optimization↗

Scaling Semantic Graph Databases in Size and Performance

In this paper we present SGEM, a full software system for accelerating large-scale semantic graph databases on commodity clusters. Unlike current approaches, SGEM addresses semantic graph databases by only employing graph methods at all the levels of the stack. On one hand, this allows exploiting the space efficiency of graph data structures and the inherent parallelism of graph algorithms. These features adapt well to the increasing system memory and core counts of modern commodity clusters. On the other hand, however, these systems are optimized for regular computation and batched data transfers, while graph methods usually are irregular and generate fine-grained data accesses with poor spatial and temporal locality. Our framework comprises a SPARQL to data parallel C compiler, a library of parallel graph methods and a custom, multithreaded runtime system. We introduce our stack, motivate its advantages with respect to other solutions and show how we solved the challenges posed by irregular behaviors. We present the result of our software stack on the Berlin SPARQL benchmarks with datasets up to 10 billion triples (a triple corresponds to a graph edge), demonstrating scaling in dataset size and in performance as more nodes are added to the cluster.

Aggregation↗

Enabling Graph Appliance for Genome Assembly

In recent years, there has been a huge growth in the amount of genomic data available as reads generated from various genome sequencers. The number of reads generated can be huge, ranging from hundreds to billions of nucleotide, each varying in size. Assembling such large amounts of data is one of the challenging computational problems for both biomedical and data scientists. Most of the genome assemblers developed have used de Bruijn graph techniques. A de Bruijn graph represents a collection of read sequences by billions of vertices and edges, which require large amounts of memory and computational power to store and process. This is the major drawback to de Bruijn graph assembly. Massively parallel, multi-threaded, shared memory systems can be leveraged to overcome some of these issues. The objective of our research is to investigate the feasibility and scalability issues of de Bruijn graph assembly on Cray s Urika-GD system; Urika-GD is a high performance graph appliance with a large shared memory and massively multithreaded custom processor designed for executing SPARQL queries over large-scale RDF data sets. However, to the best of our knowledge, there is no research on representing a de Bruijn graph as an RDF graph or finding Eulerian paths in RDF graphs using SPARQL for potential genome discovery. In this paper, we address the issues involved in representing a de Bruin graphs as RDF graphs and propose an iterative querying approach for finding Eulerian paths in large RDF graphs. We evaluate the performance of our implementation on real world ebola genome datasets and illustrate how genome assembly can be accomplished with Urika-GD using iterative SPARQL queries.

Big data↗

Traleika Glacier: A hardware-software co-designed approach to exascale computing

The move from current petascale machines to future exascale machines will need both hardware improvements and software changes. Hardware will need to evolve to focus primarily on features that lower energy consumption: near-threshold voltage operation, fine-grained power and clock management and heterogeneity. Software will also need to evolve and be able to express more parallelism, become more dynamic and adaptable in order to be able to operate on a much more variable hardware. Here, we present Traleika Glacier, an effort that seeks to evaluate radical design changes to meet the constraints, both in terms of power and cost, of exascale computing. The salient features of the hardware design presented in the work include a) a use of heterogeneous cores, b) a redesign of the memory system that centers around hierarchical scratchpads and a global address space, c) the hardware acceleration of certain memory and network operations through specialized engines and, d) very fine-grained control and monitoring capabilities. On the software side, we describe a task-based runtime system, the Open Community Runtime (OCR) which aims to express a wide range of higher-level programming models with a very limited set of core concepts: event-driven tasks for computation, events for synchronization and relocatable data-blocks for data management.

97 MATHEMATICS AND COMPUTING↗

Final Report for Project DE-FC02-06ER25755 [Pmodels2]

In this report, we describe the research accomplished by the OSU team under the Pmodels2 project. The team has worked on various angles: designing high performance MPI implementations on modern networking technologies (Mellanox InfiniBand (including the new ConnectX2 architecture and Quad Data Rate), QLogic InfiniPath, the emerging 10GigE/iWARP and RDMA over Converged Enhanced Ethernet (RoCE) and Obsidian IB-WAN), studying MPI scalability issues for multi-thousand node clusters using XRC transport, scalable job start-up, dynamic process management support, efficient one-sided communication, protocol offloading and designing scalable collective communication libraries for emerging multi-core architectures. New designs conforming to the Argonne’s Nemesis interface have also been carried out. All of these above solutions have been integrated into the open-source MVAPICH/MVAPICH2 software. This software is currently being used by more than 2,100 organizations worldwide (in 71 countries). As of January ’14, more than 200,000 downloads have taken place from the OSU Web site. In addition, many InfiniBand vendors, server vendors, system integrators and Linux distributors have been incorporating MVAPICH/MVAPICH2 into their software stacks and distributing it. Several InfiniBand systems using MVAPICH/MVAPICH2 have obtained positions in the TOP500 ranking of supercomputers in the world. The latest November ’13 ranking include the following systems: 7th ranked Stampede system at TACC with 462,462 cores; 11th ranked Tsubame 2.5 system at Tokyo Institute of Technology with 74,358 cores; 16th ranked Pleiades system at NASA with 81,920 cores; Work on PGAS models has proceeded on multiple directions. The Scioto framework, which supports task-parallelism in one-sided and global-view parallel programming, has been extended to allow multi-processor tasks that are executed by processor groups. A quantum Monte Carlo application is being ported onto the extended Scioto framework. A public release of Global Trees (GT) has been made, along with the Global Chunks (GC) framework on which GT is built. The Global Chunks (GC) layer is also being used as the basis for the development of a higher level Global Graphs (GG) layer. The Global Graphs (GG) system will provide a global address space view of distributed graph data structures on distributed memory systems.

97 MATHEMATICS AND COMPUTING↗

Enabling the First Ever Measurement of Coherent Neutrino Scattering Through Background Neutron Measurements.

Using High Performance Computing to Examine the Processes of Neurogenesis Underlying Pattern Separation/Completion of Episodic Information - Sandia researchers developed novel methods and metrics for studying the computational function of neurogenesis,thus generating substantial impact to the neuroscience and neural computing communities. This work could benefit applications in machine learning and other analysis activities. The purpose of this project was to computationally model the impact of neural population dynamics within the neurobiological memory system in order to examine how subareas in the brain enable pattern separation and completion of information in memory across time as associated experiences.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Distributed memory, GPU accelerated Fock construction for hybrid, Gaussian basis density functional theory

With the growing reliance of modern supercomputers on accelerator-based architecture such a graphics processing units (GPUs), the development and optimization of electronic structure methods to exploit these massively parallel resources has become a recent priority. While significant strides have been made in the development GPU accelerated, distributed memory algorithms for many modern electronic structure methods, the primary focus of GPU development for Gaussian basis atomic orbital methods has been for shared memory systems with only a handful of examples pursing massive parallelism. Here in this work, we present a set of distributed memory algorithms for the evaluation of the Coulomb and exact exchange matrices for hybrid Kohn–Sham DFT with Gaussian basis sets via direct density-fitted (DF-J-Engine) and seminumerical (sn-K) methods, respectively. The absolute performance and strong scalability of the developed methods are demonstrated on systems ranging from a few hundred to over one thousand atoms using up to 128 NVIDIA A100 GPUs on the Perlmutter supercomputer.

97 MATHEMATICS AND COMPUTING↗

Neuroplastic Response After Radiation Therapy for Pediatric Brain Tumors: A Pilot Study

Purpose: Clinically effective measurement of cognitive toxicity from photon radiation therapy (XRT) should be accurate, sensitive, and specific. This pilot study tested translational findings on phasic changes in children's memory systems that are sensitive and insensitive to toxic XRT effects to identify a possible neuroplastic effect. Methods and Materials: Memory processes were prospectively tested before XRT and at 3 later time points up to 2 years in 35 children with mixed primary brain tumors who had not experienced recurrence. Memory processes were verbal-semantic, visual-semantic, and visual-perceptual, including accuracy, speed to recall, encoding, retrieval, and recognition. The mixed-effects model included time (to estimate slope), covariates (age, tumor locus, XRT field, and medications) as fixed effects, and individual random intercepts. A sensitivity analysis examined the influence of XRT dose to the hippocampi on memory. Results: Retrieval from long-term verbal-semantic memory declined 2 months after completing XRT, as seen in adults, and was lowest at 1 year, which was delayed in comparison with adults. Double dissociation from visual-perceptual memory at baseline and 2 months was found, consistent with adults. Recovery was demonstrated 2 years after XRT. Patterns were unchanged when dose to hippocampus was included in the model. Conclusions: Verbal and semantic long-term retrieval is specifically sensitive to XRT-related cognitive dysfunction, without effect on visual-perceptual memory. Children reached nadir in XRT-sensitive memory 1 year after XRT and recovered by 2 years, which is later than that observed in adults. The protracted period of post-XRT injury may represent the maturation of the human hippocampus and white matter into late adolescence.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

A performance model for the communication in fast multipole methods on high-performance computing platforms

Exascale systems are predicted to have approximately 1 billion cores, assuming gigahertz cores. Limitations on affordable network topologies for distributed memory systems of such massive scale bring new challenges to the currently dominant parallel programing model. Currently, there are many efforts to evaluate the hardware and software bottlenecks of exascale designs. It is therefore of interest to model application performance and to understand what changes need to be made to ensure extrapolated scalability. The fast multipole method (FMM) was originally developed for accelerating N-body problems in astrophysics and molecular dynamics but has recently been extended to a wider range of problems. Its high arithmetic intensity combined with its linear complexity and asynchronous communication patterns make it a promising algorithm for exascale systems. In this paper, we discuss the challenges for FMM on current parallel computers and future exascale architectures, with a focus on internode communication. We focus on the communication part only; the efficiency of the computational kernels are beyond the scope of the present study. We develop a performance model that considers the communication patterns of the FMM and observe a good match between our model and the actual communication time on four high-performance computing (HPC) systems, when latency, bandwidth, network topology, and multicore penalties are all taken into account. To our knowledge, this is the first formal characterization of internode communication in FMM that validates the model against actual measurements of communication time. The ultimate communication model is predictive in an absolute sense; however, on complex systems, this objective is often out of reach or of a difficulty out of proportion to its benefit when there exists a simpler model that is inexpensive and sufficient to guide coding decisions leading to improved scaling. The current model provides such guidance.

Computer Science↗

Unveiling the nature of Ga-based chalcogenides for electrical switching selectors

Three-dimensional phase-change memory with stackable crossbar architecture is a promising technology to meet the urgent demands for high-density storage and rapid information processing in the era of explosive data growth. The performance depends strongly on the properties of ovonic threshold switching (OTS) selectors, which control the on/off states of memory units. Amorphous GaS serves as an outstanding OTS material, distinguished by its sizable mobility gap and high crystallization temperature, while the underlying mechanism continues to be inadequately comprehended. Here, in this work, we systematically studied the structural and electronic properties of amorphous Ga-X (X = S/Se/Te) using first-principles calculations. The results show that Ga atoms adopt tetrahedral motifs, while S/Se/Te atoms predominantly exhibit the structure of a distorted triangular pyramid. This structural arrangement is ascribed to the substantial dative bonds formed by the lone-pair electrons of the anions and the vacant sp3 orbitals around Ga atoms. Large mobility gaps (e.g., GaS: 2.43 eV, GaSe: 1.76 eV, GaTe: 1.26 eV) and distinct mid-gap states (e.g., ∼0.66 eV above valence band tail) ensure that these three chalcogenide glasses can be switched on under an external electric field while effectively suppressing leakage current without a bias, and the defect electronic states originate from short, robust Ga-Ga bonds due to the formation of distorted chain-like local structures. Our research elucidates the mechanisms of amorphous Ga-X as OTS materials, enriching the spectrum of electrical switching selectors by incorporating III-VI chalcogenides. This inclusion offers novel opportunities for the refinement and optimization of high-density integrated memory systems.

36 MATERIALS SCIENCE↗

Formal Definitions and Performance Comparison of Consistency Models for Parallel File Systems

The semantics of HPC storage systems are defined by the consistency models to which they abide. Storage consistency models have been less studied than their counterparts in memory systems, with the exception of the POSIX standard and its strict consistency model. The use of POSIX consistency imposes a performance penalty that becomes more significant as the scale of parallel file systems increases and the access time to storage devices, such as node-local solid storage devices, decreases. While some efforts have been made to adopt relaxed storage consistency models, these models are often defined informally and ambiguously as by-products of a particular implementation. Here in this work, we establish a connection between memory consistency models and storage consistency models and revisit the key design choices of storage consistency models from a high-level perspective. Further, we propose a formal and unified framework for defining storage consistency models and a layered implementation that can be used to easily evaluate their relative performance for different I/O workloads. Finally, we conduct a comprehensive performance comparison of two relaxed consistency models on a range of commonly seen parallel I/O workloads, such as checkpoint/restart of scientific applications and random reads of deep learning applications. We demonstrate that for certain I/O scenarios, a weaker consistency model can significantly improve the I/O performance. For instance, in small random reads that are typically found in deep learning applications, session consistency achieved a 5x improvement in I/O bandwidth compared to commit consistency, even at small scales.

97 MATHEMATICS AND COMPUTING↗

On the use of flux limiters in the discrete ordinates method for 3D radiation calculations in absorbing and scattering media

The application of flux limiters to the discrete ordinates method (DOM), S{sub N}, for radiative transfer calculations is discussed and analyzed for 3D enclosures for cases in which the intensities are strongly coupled to each other such as: radiative equilibrium and scattering media. A Newton-Krylov iterative method (GMRES) solves the final systems of linear equations along with a domain decomposition strategy for parallel computation using message passing libraries in a distributed memory system. Ray effects due to angular discretization and errors due to domain decomposition are minimized until small variations are introduced by these effects in order to focus on the influence of flux limiters on errors due to spatial discretization, known as numerical diffusion, smearing or false scattering. Results are presented for the DOM-integrated quantities such as heat flux, irradiation and emission. A variety of flux limiters are compared to 'exact' solutions available in the literature, such as the integral solution of the RTE for pure absorbing-emitting media and isotropic scattering cases and a Monte Carlo solution for a forward scattering case. Additionally, a non-homogeneous 3D enclosure is included to extend the use of flux limiters to more practical cases. The overall balance of convergence, accuracy, speed and stability using flux limiters is shown to be superior compared to step schemes for any test case.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Kinetics of the xanthophyll cycle and its role in photoprotective memory and response

Efficiently balancing photochemistry and photoprotection is crucial for survival and productivity of photosynthetic organisms in the rapidly fluctuating light levels found in natural environments. The ability to respond quickly to sudden changes in light level is clearly advantageous. In the alga Nannochloropsis oceanica we observed an ability to respond rapidly to sudden increases in light level which occur soon after a previous high-light exposure. This ability implies a kind of memory. In this work, we explore the xanthophyll cycle in N. oceanica as a short-term photoprotective memory system. By combining snapshot fluorescence lifetime measurements with a biochemistry-based quantitative model, we show that short-term memory arises from the xanthophyll cycle. In addition, the model enables us to characterize the relative quenching abilities of the three xanthophyll cycle components. Given the ubiquity of the xanthophyll cycle in photosynthetic organisms the model described here will be of utility in improving our understanding of vascular plant and algal photoprotection with important implications for crop productivity.

59 BASIC BIOLOGICAL SCIENCES↗

Blackcomb2: Hardware-Software Co-design for Nonvolatile Memory in Exascale Systems

The impending shift from DRAM and Flash to fast nonvolatile memory (NVM) provides a game-changing opportunity to rethink traditional system architectures; to address their energy inefficiencies; to leverage the new devices for greater performance; and to exploit the new capability of nonvolatility to enhance system resilience, even in the face of larger system scale and degraded reliability of components. Building on the accomplishments of the Blackcomb Project, funded in 2010, we propose to identify, evaluate, and optimize the most promising NVM hardware and software technologies, which are essential to provide the necessary memory capacity, performance, resilience, and energy efficiency in exascale systems. Capacity and energy are the key drivers. Today's pressing application requirements mandate exascale computing; the datasets and concurrency of these applications demand memory capacity that cannot be met by traditional memory technologies. Although charge-based DRAM will endure, experts foresee only modest DRAM capacity gains. And, as moving data between levels of the hierarchy, and storing and accessing data on disk consumes significant energy, a more energy efficient solution is required. We, therefore, posit that exascale systems will need high density, energy-efficient storage technologies, such as nonvolatile memory for access, transformation, and management of exascale data. To address these issues, in this renewal request, we propose continue the work of our vertically- integrated, focussed team, where our co-design is informed by proxies and interactions with DOE co-design centers. In software, we start by assuming that nonvolatility requires new abstractions and new infrastructure, allowing persistent objects, and delivering performance gains for _le systems and robustness through fast checkpointing. In architecture, we evaluate the tradeoffs of performance and energy by employing a scalable and accurate simulation system to analyze realistic proxy applications. Through this co-design process, we rethink the whole memory architecture, from the cell to the array to the device, with ECC, wearout management, and interfaces. Finally, we provide open-source software models of these NVM-based memory systems.

97 MATHEMATICS AND COMPUTING↗