Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed Memory Parallel”

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 199 records · Page 11

Block-Parallel Data Analysis with DIY2

DIY2 is a programming model and runtime for block-parallel analytics on distributed-memory machines. Its main abstraction is block-structured data parallelism: data are decomposed into blocks; blocks are assigned to processing elements (processes or threads); computation is described as iterations over these blocks, and communication between blocks is defined by reusable patterns. By expressing computation in this general form, the DIY2 runtime is free to optimize the movement of blocks between slow and fast memories (disk and flash vs. DRAM) and to concurrently execute blocks residing in memory with multiple threads. This enables the same program to execute in-core, out-of-core, serial, parallel, single-threaded, multithreaded, or combinations thereof. This paper describes the implementation of the main features of the DIY2 programming model and optimizations to improve performance. DIY2 is evaluated on benchmark test cases to establish baseline performance for several common patterns and on larger complete analysis codes running on large-scale HPC machines.

97 MATHEMATICS AND COMPUTING↗

Global Arrays Parallel Programming Toolkit

The two predominant classes of programming models for parallel computing are distributed memory and shared memory. Both shared memory and distributed memory models have advantages and shortcomings. Shared memory model is much easier to use but it ignores data locality/placement. Given the hierarchical nature of the memory subsystems in modern computers this characteristic can have a negative impact on performance and scalability. Careful code restructuring to increase data reuse and replacing fine grain load/stores with block access to shared data can address the problem and yield performance for shared memory that is competitive with message-passing. However, this performance comes at the cost of compromising the ease of use that the shared memory model advertises. Distributed memory models, such as message-passing or one-sided communication, offer performance and scalability but they are difficult to program. The Global Arrays toolkit attempts to offer the best features of both models. It implements a shared-memory programming model in which data locality is managed by the programmer. This management is achieved by calls to functions that transfer data between a global address space (a distributed array) and local storage. In this respect, the GA model has similarities to the distributed shared-memory models that provide an explicit acquire/release protocol. However, the GA model acknowledges that remote data is slower to access than local data and allows data locality to be specified by the programmer and hence managed. GA is related to the global address space languages such as UPC, Titanium, and, to a lesser extent, Co-Array Fortran. In addition, by providing a set of data-parallel operations, GA is also related to data-parallel languages such as HPF, ZPL, and Data Parallel C. However, the Global Array programming model is implemented as a library that works with most languages used for technical computing and does not rely on compiler technology for achieving parallel efficiency. It also supports a combination of task- and data-parallelism and is available as an extension of the message passing (MPI) model. The GA model exposes to the programmer the hierarchical memory of modern high-performance computer systems, and by recognizing the communication overhead for remote data transfer, it promotes data reuse and locality of reference. Virtually all the scalable architectures possess non-uniform memory access characteristics that reflect their multi-level memory hierarchies. These hierarchies typically comprise processor registers, multiple levels of cache, local memory, and remote memory. Over time, both the number of levels and the cost (in processor cycles) of accessing deeper levels has been increasing. It is important for any scalable programming model to address memory hierarchy since it is critical to the efficient execution of scalable applications.

ARMCI↗

Scalable Distributed Memory Community Detection Using Vite

Graph clustering, popularly known as community detection, is a fundamental graph operation used in several applications of relevance for network analysis and cybersecurity. Community detection is an operation of partitioning a network into an arbitrary number of “communities” such that each com munity represents a tightly-knit group of nodes with relatively sparser connections to the rest of the nodes in the network. The need to compute clustering on large-scale networks necessitates the development of efficient parallel algorithms capable of exploiting modern parallel architectures. However, due to their irregular and inherently sequential nature, many of the current algorithms for community detection are challenging to parallelize. In response to the 2018 Graph Challenge, we present Vite which is a distributed-memory parallel community detection method. Vite is a parallelization of the algorithmic template provided by a widely-used serial method (Louvain). Alongside, it uses several heuristics aimed at significantly improving parallelism and performance, while preserving output quality. Using the inputs from the 2018 Graph Challenge (static and streaming), we demonstrate superior performance and high quality solutions based on four parallelization heuristics.

community detection↗

PaKman: Scalable Assembly of Large Genomes on Distributed Memory Machines

De novo genome assembly is a fundamental problem in the field of bioinformatics, that aims to assemble the DNA sequence of an unknown genome from numerous short DNA fragments (aka reads) obtained from it. With the advent of high-throughput sequencing technologies, billions of reads can be generated in a matter of hours, necessitating efficient parallelization of the assembly process. While multiple parallel solutions have been proposed in the past, conducting a large- scale assembly at scale remains a challenging problem because of the inherent complexities associated with data movement, and irregular access footprints of memory and I/O operations. In this paper, we present a novel algorithm, called PaKman, to address the problem of performing large-scale genome assemblies on a distributed memory parallel computer. Our approach focuses on improving performance through a combination of novel data structures and algorithmic strategies for reducing the communication and I/O footprint during the assembly process. A key aspect of our algorithm is its graph data structure, which comprises fat nodes (or what we call "macro-nodes") that reduce the communication burden during contig generation. We present an extensive performance and qualitative evaluation of our algorithm, including comparisons to other state-of-the-art parallel assemblers. Our results demonstrate the ability to achieve near-linear speedups on up to 8K cores (tested); outperform state-of-the-art distributed memory and shared memory tools in performance while delivering comparable (if not better) quality; and reduce time to solution significantly. For instance, PaKman is able to complete an assembly of the full human genome in just over a minute on 8K cores.

Ghosh, Priyanka↗

Towards Scalable Optimal Sequence Homology Detection

Abstract—The field of bioinformatics and computational biol- ogy is experiencing a data revolution — experimental techniques to procure data have increased in throughput, improved in accuracy and reduced in costs. This has spurred an array of high profile sequencing and data generation projects. While the data repositories represent untapped reservoirs of rich information critical for scientific breakthroughs, the analytical software tools that are needed to analyze large volumes of such sequence data have significantly lagged behind in their capacity to scale. In this paper, we address homology detection, which is a funda- mental problem in large-scale sequence analysis with numerous applications. We present a scalable framework to conduct large- scale optimal homology detection on massively parallel super- computing platforms. Our approach employs distributed memory work stealing to effectively parallelize optimal pairwise alignment computation tasks. Results on 120,000 cores of the Hopper Cray XE6 supercomputer demonstrate strong scaling and up to 2.42 × 107 optimal pairwise sequence alignments computed per second (PSAPS), the highest reported in the literature.

Daily, Jeffrey A.↗

Frameworks, Algorithms, and Scalable Technologies for Mathematics (FASTMath) SciDAC Institute

The FASTMath SciDAC Institute addressed two key challenges that application scientists faced at the beginning of SciDAC-3. First, FASTMath helped them continue to improve the quality of their simulations by increasing accuracy and reliability of both their software and algorithms. Second, FASTMath helped them adapt their computations to make effective use of high-end computing facilities acquired by DOE over the past five years. This required the development of new mathematical algorithms that were appropriate for the physics problems being solved, implementations that scaled to million-way parallelism, and software that effectively leveraged both distributed memory and on-node parallelism. The Rensselaer Polytechnic Institute team’s efforts focused on the unstructured mesh technologies developed by FASTMath.

97 MATHEMATICS AND COMPUTING↗

Parallel streaming between heterogeneous HPC resources for real-time analysis

Performing analysis or generating visualizations concurrently with high performance simulations can yield great benefits compared to post-processing data. Writing and reading large volumes of data can be reduced or eliminated, thereby producing an I/O cost savings. One such method for concurrent simulation and analysis is in transit - streaming data from the resource running the simulation to a separate resource running the analysis. In transit analysis can be beneficial since computational resources may not have certain resources needed for visualization and analysis (e.g. GPUs) and to reduce the impact of performing analysis tasks to the run time of the simulation. When sending and receiving data in transit, data redistribution mechanisms are needed in order to support heterogeneous data layouts that may be required by the simulation and analysis applications. The work described in this paper compares two mechanisms for on-the-fly data redistribution when streaming data in parallel between two distributed memory applications. Finally, our results show that it is often more advantageous to stream data in the same layout as the sender and redistribute data amongst processes on the receiving end than to stream data in the final layout needed by the receiver.

97 MATHEMATICS AND COMPUTING↗

Performance upgrades to the MCNP6 burnup capability for large scale depletion calculations

The first MCNP based inline Monte Carlo depletion capability was officially released from the Radiation Safety Information and Computational Center as MCNPX 2.6.0. With the merger of MCNPX and MCNP5, MCNP6 combined the capability of both simulation tools, as well as providing new advanced technology, in a single radiation transport code. The new MCNP6 depletion capability was first showcased at the International Congress for Advancements in Nuclear Power Plants (ICAPP) meeting in 2012. At that conference the new capabilities addressed included the combined distributive and shared memory parallel architecture for the burnup capability, improved memory management, physics enhancements, and new predictability as compared to the H.B Robinson Benchmark. At Los Alamos National Laboratory, a special purpose cluster named “tebow,” was constructed such to maximize available RAM per CPU, as well as leveraging swap space with solid state hard drives, to allow larger scale depletion calculations (allowing for significantly more burnable regions than previously examined). As the MCNP6 burnup capability was scaled to larger numbers of burnable regions, a noticeable slowdown was realized.This paper details two specific computational performance strategies for improving calculation speedup: (1) retrieving cross sections during transport; and (2) tallying mechanisms specific to burnup in MCNP. To combat this slowdown new performance upgrades were developed and integrated into MCNP6 1.2.

42 ENGINEERING↗

OpenSHMEM over MPI-3

OpenSHMEM is a community effort to unifyt and standardize the SHMEM programming model. MPI (Message Passing Interface) is a well-known community standard for parallel programming using distributed memory. The most recen t release of MPI, version 3.0, was designed in part to support programming models like SHMEM.OSHMPI is an implementation of the OpenSHMEM standard using MPI-3 for the Linux operating system. It is the first implementation of SHMEM over MPI one-sided communication and has the potential to be widely adopted due to the portability and widely availability of Linux and MPI-3. OSHMPI has been tested on a variety of systems and implementations of MPI-3, includingInfiniBand clusters using MVAPICH2 and SGI shared-memory supercomputers using MPICH. Current support is limited to Linux but may be extended to Apple OSX if there is sufficient interest. The code is opensource via https://github.com/jeffhammond/oshmpi

Sayan Ghosh, JeffHammond↗

Asynchronous Navier-Stokes Solver on 3D Unstructured Grids for the Exascale Era

This project has developed multiple fluid dynamics solvers for complex 3D flows using fully asynchronous distributed-memory task-parallel algorithms on top of the Charm++ runtime system. The algorithms solve the Euler or Navier-Stokes equations of compressible flows using unstructured tetrahedron meshes with optional solution-adaptive mesh-, and polynomial-order refinement. We have demonstrated excellent strong scaling up 50K compute cores and the benefits of Charm++’s automatic load balancing.

97 MATHEMATICS AND COMPUTING↗

OpenSHMEM over MPI-3

OpenSHMEM is a community effort to unifyt and standardize the SHMEM programming model. MPI (Message Passing Interface) is a well-known community standard for parallel programming using distributed memory. The most recen t release of MPI, version 3.0, was designed in part to support programming models like SHMEM.OSHMPI is an implementation of the OpenSHMEM standard using MPI-3 for the Linux operating system. It is the first implementation of SHMEM over MPI one-sided communication and has the potential to be widely adopted due to the portability and widely availability of Linux and MPI-3. OSHMPI has been tested on a variety of systems and implementations of MPI-3, includingInfiniBand clusters using MVAPICH2 and SGI shared-memory supercomputers using MPICH. Current support is limited to Linux but may be extended to Apple OSX if there is sufficient interest. The code is opensource via https://github.com/jeffhammond/oshmpi

Sayan Ghosh, Jeff Hammond↗

An Adaptive-Mesh-Refinement Based Computational Tool for Simulating Catalysis at Mesoscale

In this work, we present a computational tool for mesoscale applications using open-source exascale- computing compatible adaptive-mesh-refinement (AMR) library, AMReX [2]. AMReX is software library that enables development of application solvers with block-structured Cartesian AMR. Our tool has capabilities to include realistic geometry representation, chemical species transport, reactions and thermodynamics that are critical for capturing mesoscale physics. A significant achievement is the ability of our solver to automatically import electron microscopy data in the form of a stereolithography (STL) or pixelated file format (mrc, tiff) without undergoing the tedious task of unstructured mesh generation. This feature allows for rapid simulation of catalyst particles with complex morphologies using an immersed-boundary formulation. The use of AMR allows for higher resolutions at catalyst surface interfaces, which in turn provides an accurate description of surface reactions and transport. Our solver uses a hybrid distributed and shared memory parallelism (OpenMP/GPU-based) with which strong scaling up to 10,000 processors for realistic catalyst particle simulations have been demonstrated.

BIOMASS FUELS,MATHEMATICS AND COMPUTING↗

Towards scaling community detection on distributed-memory heterogeneous systems

Distributed multi-GPU systems pose significant challenges and opportunities for efficient execution of parallel applications. Graph algorithms are generally characterized by irregular memory accesses, low computation to communication ratios, and load balancing problems that are especially hard to address on multi-GPU systems. Graph community detection is an important problem in the emerging domain of graph analytics with numerous applications. In this paper, we present our ongoing work on distributed-memory multi-GPU implementation for graph community detection. Our work parallelizes the widely used (albeit serial) Louvain method on distributed multi-GPU platforms. Supported by an extensive set of experiments on a multi-GPU enabled supercomputer (OLCF Summit) and a single compute node (Nvidia DGX-2®), we demonstrate competitive performance to existing distributed-memory CPU-based implementation, and up to 6.5 better results than Nvidia RAPIDS® CUGRAPH. To the best of our knowledge, this work represents the first effort for community detection on distributed multi-GPU systems. Our approach and related findings can be extended to numerous other iterative graph algorithms on multi-GPU systems.

97 MATHEMATICS AND COMPUTING↗

Support for non-locking parallel reception of packets belonging to a single memory reception FIFO

A method and apparatus for distributed parallel messaging in a parallel computing system. A plurality of DMA engine units are configured in a multiprocessor system to operate in parallel, one DMA engine unit for transferring a current packet received at a network reception queue to a memory location in a memory FIFO (rmFIFO) region of a memory. A control unit implements logic to determine whether any prior received packet destined for that rmFIFO is still in a process of being stored in the associated memory by another DMA engine unit of the plurality, and prevent the one DMA engine unit from indicating completion of storing the current received packet in the reception memory FIFO (rmFIFO) until all prior received packets destined for that rmFIFO are completely stored by the other DMA engine units. Thus, there is provided non-locking support so that multiple packets destined for a single rmFIFO are transferred and stored in parallel to predetermined locations in a memory.

97 MATHEMATICS AND COMPUTING↗

Devastator Parallel Discrete Event Simulation Runtime (Devastator) v1.0

The Devastator runtime is a modern C++ implementation of optimistic parallel discrete event simulation methods. Devastator allows simulation application code to productively specify their component and event functionality with C++14 constructs. It utilizes GASNet-EX for distributed memory communication and includes parallel performance optimizations such as light-weight thread message queues and asynchronous GVT. Furthermore, it supports efficient event broadcasts and pause-rewind-resume functionality to support periodic load balancing and outer loop optimization algorithms.

Chan, Cy↗