Generic, Hierarchical Framework for Massively Parallel Wang-Landau Sampling
Not Available
SEARCH · Search NASA
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.
Not Available
This summer internship consisted of learning computer programming skills and languages including Thrust, CUDA, C++, C and Linux. The application of this programming is for use in GPU, graphics processing units, that are processors that perform math intensive calculations rapidly in three dimensions typically in advanced video games (“NVIDIA GeForce…”). Our goal is to apply this technology towards exa-scale computing on the LANL supercomputing infrastructure. My mentors use these machines to do computations. The completed algorithms include KD tree and Top K. In order to access the GPU, CUDA was used. CUDA is a parallel computer architecture for general computing on the GPUs. There are different means by which to access CUDA. The first approach uses libraries, like Thrust, Linear Algebra, and Signal and Image Processing. Another approach parallelizes loops in Fortran or C and in the last approach people develop custom parallel algorithms using familiar language (“What is CUDA”). The algorithms I worked on this summer use the first approach of using the thrust library of parallel algorithms and data structures.
In this work, we describe the capabilities of Magnonic Holographic Memory (MHM) for parallel database search and prime factorization. MHM is a type of holographic device, which utilizes spin waves for data transfer and processing. Its operation is based on the correlation between the phases and the amplitudes of the input spin waves and the output inductive voltage. The input of MHM is provided by the phased array of spin wave generating elements allowing the producing of phase patterns of an arbitrary form. The latter makes it possible to code logic states into the phases of propagating waves and exploit wave superposition for parallel data processing. We present the results of numerical modeling illustrating parallel database search and prime factorization. The results of numerical simulations on the database search are in agreement with the available experimental data. The use of classical wave interference may results in a significant speedup over the conventional digital logic circuits in special task data processing (e.g., √n in database search). Potentially, magnonic holographic devices can be implemented as complementary logic units to digital processors. Physical limitations and technological constrains of the spin wave approach are also discussed.
The recent emergence of hardware architectures characterized by many-core or accelerated processors has opened new opportunities for concurrent programming models taking advantage of both SIMD and SIMT architectures. The GeantV vector prototype for detector simulations has been designed to exploit both the vector capability of mainstream CPUs and multi-threading capabilities of coprocessors including NVidia GPUs and Intel Xeon Phi. The characteristics of these architectures are very different in terms of the vectorization depth, parallelization needed to achieve optimal performance or memory access latency and speed. An additional challenge is to avoid the code duplication often inherent to supporting heterogeneous platforms. In this paper we present the first experience of vectorizing electromagnetic physics models developed for the GeantV project.
As multi-petascale and exa-scale high-performance computing (HPC) systems inevitably have to deal with a number of resilience challenges, such as a significant growth in component count and smaller circuit sizes with lower circuit voltages, redundancy may offer an acceptable level of resilience that traditional fault tolerance techniques, such as checkpoint/restart, do not. Although redundancy in HPC is quite controversial due to the associated cost for redundant components, the constantly increasing number of cores-per-processor is tilting this cost calculation toward a system design where computation, such as for redundancy, is much cheaper and communication, needed for checkpoint/restart, is much more expensive. Recent research and development activities in redundancy for Message Passing Interface (MPI) applications focused on availability/reliability models and replication algorithms. This paper takes a first step toward solving an open research problem associated with running a parallel application redundantly, which is file I/O under redundancy. The approach intercepts file I/O calls made by a redundant application to employ coordination protocols that execute file I/O operations in a redundancy-oblivious fashion when accessing a node-local file system, or in a redundancy-aware fashion when accessing a shared networked file system. A proof-of concept prototype is presented and a number of coordination protocols are described and evaluated. The results show the performance impact for redundantly accessing a shared networked file system, but also demonstrate the capability to regain performance by utilizing MPI communication between replicas and parallel file I/O.
FLASH is a multiphysics multiscale adaptive mesh refinement (AMR) code originally designed for simulation of reactive flows often found in Astrophysics. With its wide user base and flexible applications configuration capability, FLASH has a dual task of maintaining scalability and portability in all its solvers. The scalability of fully explicit solvers in the code is tied very closely to that of the underlying mesh. Others such as the Poisson solver based on a multigrid method have more complex scaling behavior. Multigrid methods suffer from processor starvation and dominating communication costs at coarser grids with increase in the number of processors. In this paper, we propose a combination of uniform grid mesh with AMR mesh, and the merger of two different sets of solvers to overcome the scalability limitation of the Poisson solver in FLASH. The principal challenge in the proposed merger is the efficiency of the communication algorithm to map the mesh back and forth between uniform grid and AMR. We present two different parallel mapping algorithms and also discuss results from performance studies of the two implementations. Copyright © 2012 John Wiley & Sons, Ltd.
Remote DMA (RDMA) enables high performance networks to reduce data copying between an application and the operating system (OS). However RDMA operations in some high performance networks require communication memory explicitly registered with the network adapter and pinned by the OS. Memory registration and pinning limits the flexibility of the memory system and reduces the amount of memory that user processes can allocate. These issues become more significant on multicore platforms, since registered memory demand grows linearly with the number of processor cores. In this paper we propose a new memory registration/deregistration strategy to reduce registered memory on multicore architectures for HPC applications. We hide the cost of dynamic memory management by offloading all dynamic memory registration and deregistration requests to a dedicated memory management helper thread. We investigate design policies and performance implications of the helper thread approach. We evaluate our framework with the NAS parallel benchmarks, for which our registration scheme significantly reduces the registered memory (23.62% on average and up to 49.39%) and avoids memory registration/deregistration costs for reused communication memory. We show that our system enables the execution of problem sizes that could not complete under existing memory registration strategies.
With an average herd size of 113 mature cows, Cayuga County is home to 280 dairy farms and 31,500 dairy milking cows producing approximately 855 million gallons of milk per year. The Cayuga Dairy industry is a major contributor to the countys economy, employing nearly 1200 people, while generating $140,000,000 of revenue from sale of milk alone. At the same time, the Cayuga County dairy industry also produces 5.7 million gallons of manure daily: (a) Nearly 34% of this manure is produced on smaller farms. (b) Digesters are expensive pieces of equipment and require attention and care. ( c) The on-farm digester systems have fairly long payback (>10 years) even for larger CAFO farms (>1000 milking cows). In 2005, Cayuga County Soil and Water Conservation District (The District), a Public Agency under Cayuga County, decided to undertake a centralized community digester project. The primary goal of the project was to develop an economically sustainable model, under the auspices of The District to address manure management issues facing the smaller dairies, improve the water quality and improve the quality of life for Cayuga County residents. It is believed that the District has accomplished this goal by completing construction of Cayuga County Regional Digester on a parcel of land behind the Cayuga County Natural Resource Center located at 7413 County House Road in the Town of Sennett in Cayuga County, New York. The digester facility consists of the following major components. 1. Transfer Station: This an indoor truck bay, where 35,000 gallons of manure from three local farms, 8,500 gallons of liquid organic food-processor waste, and 1,200 gallons of brown grease are unloaded from tanker trucks and the digested slurry is loaded onto the tanker trucks for delivery back to the participating farms. 2. Anaerobic Digester: The project utilizes a hydraulic mix anaerobic digester, a unique design that has no internal moving parts for mixing. The digester, which operates at mesophilic temperatures, is designed to process the daily feedstock and produce 220,000 SCF2 of biogas per day. The digester also produces 44,000 gallons of digested slurry per day. 3. Biogas Conditioning System: The plant employs a biological biogas conditioning system to remove the H2S and moisture contents of the biogas and prepare it to be used by the plant generation system. 4. Combined Heat and Power System (CHP): This is a 633kW high efficiency biogas-fired GE-Jenbacher model JMS-312 GS-NL reciprocating engine cogeneration system. The heat recovery system incorporated into the package is designed to capture the waste heat from the engine exhaust, the jacket cooling water and the engine oil circuit. 5. Electrical Substation and Power Distribution Systems: An electrical distribution system has been constructed on-site that aggregates the electrical service of the different county buildings on the District campus into a county owned electric distribution system that is interconnected with the CHP and the local electric grid. The electrical system is designed, in accordance with the utility guidelines, to allow grid-parallel operation of CHP and provide for import and export of electric power. 6. Thermal Energy Distribution System: The heat recovery system has been integrated into a high temperature water distribution system that distributes the heat to the thermal circuits for the anaerobic digester facility. Additional piping has also been installed to transfer the remaining thermal energy to other county buildings on the campus. On a daily basis, the plant will co-process 35,000 gallons of manure from local dairy farms, 8,500 gallons of food-processor waste and 1,200 gallons of brown grease to produce 200,000 ft3/d of biogas and 44,000 gallons of pathogen-free nutrient-rich digested slurry for agricultural use by farms and in the local area.The biogas fueled CHP produces 5,157,000 kWh of electricity and 19,506 dekatherms of thermal energy per year. Electrical power generated by the cogeneration system powers all the buildings on the Cayuga County campus and any surplus power is exported to the grid under a power purchase agreement. Heat recovered from the cogeneration system will be used to maintain the temperature of the process equipment and the excess will be transported to the Cayuga County Public Safety Building to offset purchase of fossil fuel to fuel the boilers. The majority of plant operations are unmanned and automated. However, the plant will have a small staff of well-trained personnel to coordinate the feedstock deliveries and shipments, supervise the day-to-day operation, monitor the systems and perform maintenance, maintain a safe and reliable operation and to respond to emergencies.
This is a Phase I report on a project to significantly enhance existing subsurface simulation software using leadership-class computing resources, allowing researchers to solve problems with greater speed and accuracy. Subsurface computer simulation is used for monitoring the behavior of contaminants around nuclear waste disposal and storage areas, groundwater flow, environmental remediation, carbon sequestration, methane hydrate production, and geothermal energy reservoir analysis. The Phase I project was a collaborative effort between Thunderhead Engineering (project lead and developers of a commercial pre- and post-processor for the TOUGH2 simulator) and Lawrence Berkeley National Laboratory (developers of the TOUGH2 simulator for subsurface flow). The Phase I project successfully identified the technical approaches to be implemented in Phase II.
The conjugate gradient is a prominent iterative method for solving systems of sparse linear equations. Large-scale scientific applications often utilize a conjugate gradient solver at their computational core. In this paper we present a field programmable gate array (FPGA) based implementation of a double precision, non-preconditioned, conjugate gradient solver for fmite-element or finite-difference methods. OUf work utilizes the SRC Computers, Inc. MAPStation hardware platform along with the 'Carte' software programming environment to ease the programming workload when working with the hybrid (CPUIFPGA) environment. The implementation is designed to handle large sparse matrices of up to order N x N where N <= 116,394, with up to 7 non-zero, 64-bit elements per sparse row. This implementation utilizes an optimized sparse matrix-vector multiply operation which is critical for obtaining high performance. Direct parallel implementations of loop unrolling and loop fusion are utilized to extract performance from the various vector/matrix operations. Rather than utilize the FPGA devices as function off-load accelerators, our implementation uses the FPGAs to implement the core conjugate gradient algorithm. Measured run-time performance data is presented comparing the FPGA implementation to a software-only version showing that the FPGA can outperform processors running up to 30x the clock rate. In conclusion we take a look at the new SRC-7 system and estimate the performance of this algorithm on that architecture.
By utilizing the molecular dynamics code SPaSM on Livermore's BlueGene/L architecture, consisting of 212 992 IBM PowerPC440 700 MHz processors, a molecular dynamics simulation was run with one trillion atoms. To demonstrate the practicality and future potential of such ultra large-scale simulations, the onset of the mechanical shear instability occurring in a system of Lennard-Jones particles arranged in a simple cubic lattice was simulated. The evolution of the instability was analyzed on-the-fly using the in-house developed massively parallel graphical object-rendering code MD{_}render.
We present a new methodology for utilizing all CPU cores and all GPUs on a heterogeneous multicore and multi-GPU system to support matrix computations e ciently. Our approach is able to achieve the objectives of a high degree of parallelism, minimized synchronization, minimized communication, and load balancing. Our main idea is to treat the heterogeneous system as a distributed-memory machine, and to use a heterogeneous 1-D block cyclic distribution to allocate data to the host system and GPUs to minimize communication. We have designed heterogeneous algorithms with two di erent tile sizes (one for CPU cores and the other for GPUs) to cope with processor heterogeneity. We propose an auto-tuning method to determine the best tile sizes to attain both high performance and load balancing. We have also implemented a new runtime system and applied it to the Cholesky and QR factorizations. Our experiments on a compute node with two Intel Westmere hexa-core CPUs and three Nvidia Fermi GPUs demonstrate good weak scalability, strong scalability, load balance, and e ciency of our approach.
We introduce a new iterative Krylov subspace-based eigensolver for the simulation of macromolecular motions on desktop multithreaded platforms equipped with multicore processors and, possibly, a graphics accelerator (GPU). The method consists of two stages, with the original problem first reduced into a simpler band-structured form by means of a high-performance compute-intensive procedure. This is followed by a memory-intensive but low-cost Krylov iteration, which is off-loaded to be computed on the GPU by means of an efficient data-parallel kernel. The experimental results reveal the performance of the new eigensolver. Concretely, when applied to the simulation of macromolecules with a few thousands degrees of freedom and the number of eigenpairs to be computed is small to moderate, the new solver outperforms other methods implemented as part of high-performance numerical linear algebra packages for multithreaded architectures.
Analysis and modeling of nuclear reactors can lead to memory overload for a single core processor when it comes to refined modeling. A method to solve this problem is called 'domain decomposition'. In the current work, domain decomposition algorithms for a combinatorial geometry Monte Carlo transport code are developed on the JCOGIN (J Combinatorial Geometry Monte Carlo transport INfrastructure). Tree-based decomposition and asynchronous communication of particle information between domains are described in the paper. Combination of domain decomposition and domain replication (particle parallelism) is demonstrated and compared with that of MERCURY code. A full-core reactor model is simulated to verify the domain decomposition algorithms using the Monte Carlo particle transport code JMCT (J Monte Carlo Transport Code), which has being developed on the JCOGIN infrastructure. Besides, influences of the domain decomposition algorithms to tally variances are discussed. (authors)
This work presents efficient solution techniques for radiative transfer in the smoothed particle hydrodynamics discretization. Two choices that impact efficiency are how the material and radiation energy are coupled, which determines the number of iterations needed to converge the emission source, and how the radiation diffusion equation is solved, which must be done in each iteration. The coupled material and radiation energy equations are solved using an inexact Newton iteration scheme based on nonlinear elimination, which reduces the number of Newton iterations needed to converge within each time step. During each Newton iteration, the radiation diffusion equation is solved using Krylov iterative methods with a multigrid preconditioner, which abstracts and optimizes much of the communication when running in parallel. The code is verified for an infinite medium problem, a one-dimensional Marshak wave, and a two and three-dimensional manufactured problem, and exhibits first-order convergence in time and second-order convergence in space. For these problems, the number of iterations needed to converge the inexact Newton scheme and the diffusion equation is independent of the number of spatial points and the number of processors.
An analogy is made and exploited between the recognition of visual objects and language parsing. A subset of regular languages is used to define a one-dimensional 'visual' language, in which the words are translational and scale invariant. This allows an exploration of the viewpoint invariant languages that can be solved by a network of concurrent, hierarchically connected processors. A language family is defined that is hierarchically tiling system recognizable (HREC). As inspired by nature, an algorithm is presented that constructs a cellular automaton that recognizes strings from a language in the HREC family. It is demonstrated how a language recognizer can be implemented from the cellular automaton using a feed-forward network of spiking neurons. This parser recognizes fixed-length strings from the language in parallel and as the computation is pipelined, a new string can be parsed in each new interval of time. The analogy with formal language theory allows inferences to be drawn regarding what class of objects can be recognized by visual cortex operating in purely feed-forward fashion and what class of objects requires a more complicated network architecture.
While there is no lack of performance counter tools for coarse-grained measurement of cache activity, there is a critical lack of tools for relating data layout to cache behavior to application performance. Generally, any nontrivial optimizations are either not done at all, or are done ”by hand” requiring significant time and expertise. To the best of our knowledge no tool available to users measures the latency of memory reference instructions for particular addresses and makes this information available to users in an easy-to-use and intuitive way. In this project, we worked to enable the Open|SpeedShop performance analysis tool to gather memory reference latency information for specific instructions and memory addresses, and to gather and display this information in an easy-to-use and intuitive way to aid performance analysts in identifying problematic data structures in their codes. This tool was primarily designed for use in the supercomputer domain as well as grid, cluster, cloud-based parallel e-commerce, and engineering systems and middleware. Ultimately, we envision a tool to automate optimization of application cache layout and utilization in the Open|SpeedShop performance analysis tool. To commercialize this software, we worked to develop core capabilities for gathering enhanced memory usage performance data from applications and create and apply novel methods for automatic data structure layout optimizations, tailoring the overall approach to support existing supercomputer and cluster programming models and constraints. In this Phase I project, we focused on infrastructure necessary to gather performance data and present it in an intuitive way to users. With the advent of enhanced Precise Event-Based Sampling (PEBS) counters on recent Intel processor architectures and equivalent technology on AMD processors, we are now in a position to access memory reference information for particular addresses. Prior to the introduction of PEBS counters, cache behavior could only be measured reliably in the aggregate across tens or hundreds of thousands of instructions. With the newest iteration of PEBS technology, cache events can be tied to a tuple of instruction pointer, target address (for both loads and stores), memory hierarchy, and observed latency. With this information we can now begin asking questions regarding the efficiency of not only regions of code, but how these regions interact with particular data structures and how these interactions evolve over time. In the short term, this information will be vital for performance analysts understanding and optimizing the behavior of their codes for the memory hierarchy. In the future, we can begin to ask how data layouts might be changed to improve performance and, for a particular application, what the theoretical optimal performance might be. The overall benefit to be produced by this effort was a commercial quality easy-to- use and scalable performance tool that will allow both beginner and experienced parallel programmers to automatically tune their applications for optimal cache usage. Effective use of such a tool can literally save weeks of performance tuning effort. Easy to use. With the proposed innovations, finding and fixing memory performance issues would be more automated and hide most to all of the performance engineer expertise ”under the hood” of the Open|SpeedShop performance tool. One of the biggest public benefits from the proposed innovations is that it makes performance analysis more usable to a larger group of application developers. Intuitive reporting of results. The Open|SpeedShop performance analysis tool has a rich set of intuitive, yet detailed reports for presenting performance results to application developers. Our goal was to leverage this existing technology to present the results from our memory performance addition to Open|SpeedShop. Suitable for experts as well as novices. Application performance is getting more difficult to measure as the hardware platforms they run on become more complicated. This makes life difficult for the application developer, in that they need to know more about the hardware platform, including the memory system hierarchy, in order to understand the performance of their application. Some application developers are comfortable in that sceario, while others want to do their scientific research and not have to understand all the nuances in the hardware platform they are running their application on. Our proposed innovations were aimed to support both experts and novice performance analysts. Useful in many markets. The enhancement to Open|SpeedShop would appeal to a broader market space, as it will be useful in scientific, commercial, and cloud computing environments. Our goal was to use technology developed initially at the and Lawrence Livermore National Laboratory combined with the development and commercial software experience of the Argo Navis Technologies, LLC (ANT) to form a powerful combination to delivery these objectives.
The Silicon Vertex Trigger (SVT) is a processor developed at CDF experiment to perform online fast and precise track reconstruction. SVT is made of two pipelined processors, the Associative Memory, finding low precision tracks, and the Track Fitte r, refining the track quality whith high precision fits. We will describe the architecture and the performances of a next generation track fitter, the GigaFitter, developed to reduce the degradation of the SVT efficiency due to the increasing instantaneou s luminosity. The GigaFitter reduces the track parameter reconstruction to a few clock cycles and can perform many fits in parallel, thus allowing high resolution tracking at very high rate. The core of the GigaFitter is implemented in a modern Xilinx Vir tex-5 FPGA chip, rich in powerful DSP arrays. The FPGA is housed on a mezzanine board which receives the data from a subset of the tracking detector and transfers the fitted tracks to a Pulsar motherboard for the final corrections. Instead of the current 12 boards, one per sector of the detector, the final system will be much more compact, consisting of a single GigaFitter Pulsar board equipped with four mezzanine cards receiving the data from the entire tracking detector. Moreover, the GigaFitter modular structure is adequate to scale for much better performances and is general enough to be easily adapted to future High Energy Physics (HEP) experiments and applications outside HEP.