Search NASA⌕ Search

SEARCH · Search NASA

Results for “multithreaded architecture”

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 37 records · Page 2

Parallel k-means++ for Multiple Shared-Memory Architectures

In recent years k-means++ has become a popular initialization technique for improved k-means clustering. To date, most of the work done to improve its performance has involved parallelizing algorithms that are only approximations of k-means++. In this paper we present a parallelization of the exact k-means++ algorithm, with a proof of its correctness. We develop implementations for three distinct shared-memory architectures: multicore CPU, high performance GPU, and the massively multithreaded Cray XMT platform. We demonstrate the scalability of the algorithm on each platform. In addition we present a visual approach for showing which platform performed k-means++ the fastest for varying data sizes.

data clustering↗

Implementing and Evaluating Multithreaded Triad Census Algorithms on the Cray XMT

Commonly represented as directed graphs, social networks depict relationships and behaviors among social entities such as people, groups, and organizations. Social network analysis denotes a class of mathematical and statistical methods designed to study and measure social networks. Beyond sociolo-gy, social network analysis methods are being applied to other types of data in other domains such as bioinformatics, computer networks, national security, and economics. For particular problems, the size of a social network can grow to millions of nodes and tens of millions of edges or more. In such cases, researchers could benefit from the application of social network analysis algorithms on high-performance architectures and systems. The Cray XMT is a third generation multithreaded system based on the Cray XT-3/4 platform. Like most other multithreaded architectures, the Cray XMT is designed to tolerate memory access latencies by switching context between threads. The processors maintain multiple threads of execution and util-ize hardware-based context switching to overlap the memory latency incurred by any thread with the computations from other threads. Due to its memory latency tolerance, the Cray XMT has the poten-tial of significantly improving the execution speed of irregular data-intensive applications such as those found in social network analysis. In this paper, we describe our experiences in developing and optimizing three implementations of a social network analysis method known as triadic analysis to execute on the Cray XMT. The three im-plementations possess different execution complexities, qualities, and characteristics. We evaluate how the various attributes of the codes affect their performance on the Cray XMT. We also explore the effects of different compiler options and execution strategies on the different triadic analysis im-plementations and identify general XMT programming issues and lessons learned.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Designing a unified programming model for heterogeneous machines

While high-efficiency machines are increasingly embracing heterogeneous architectures and massive multithreading, contemporary mainstream programming languages reflect a mental model in which processing elements are homogeneous, concurrency is limited, and memory is a flat undifferentiated pool of storage. Moreover, the current state of the art in programming heterogeneous machines tends towards using separate programming models, such as OpenMP and CUDA, for different portions of the machine. Both of these factors make programming emerging heterogeneous machines unnecessarily difficult. We describe the design of the Phalanx programming model, which seeks to provide a unified programming model for heterogeneous machines. It provides constructs for bulk parallelism, synchronization, and data placement which operate across the entire machine. Our prototype implementation is able to launch and coordinate work on both CPU and GPU processors within a single node, and by leveraging the GASNet runtime, is able to run across all the nodes of a distributed-memory machine.

Garland, Michael↗

SMT-Aware Instantaneous Footprint Optimization

Modern architectures employ simultaneous multithreading (SMT) to increase thread-level parallelism. SMT threads share many functional units and the whole memory hierarchy of a physical core. Without a careful code design, SMT threads can easily contend with each other for these shared resources, causing severe performance degradation. Minimizing SMT thread contention for HPC applications running on dedicated platforms is very challenging, because they usually spawn threads within Single Program Multiple Data (SPMD) models. To address this important issue, we introduce a simple scheme for SMT-aware code optimization, which aims to reduce the memory contention across SMT threads.

Instantaneous Footprint↗

The Ongoing Evolution of OpenMP

Here, this paper presents an overview of the past, present and future of the OpenMP application programming interface (API). While the API originally specified a small set of directives that guided shared memory fork-join parallelization of loops and program sections, OpenMP now provides a richer set of directives that capture a wide range of parallelization strategies that are not strictly limited to shared memory. As we look toward the future of OpenMP, we immediately see further evolution of the support for that range of parallelization strategies and the addition of direct support for debugging and performance analysis tools. Looking beyond the next major release of the specification of the OpenMP API, we expect the specification eventually to include support for more parallelization strategies and to embrace closer integration into its Fortran, C and, in particular, C++ base languages, which will likely require the API to adopt additional programming abstractions.

97 MATHEMATICS AND COMPUTING↗

System Architecture Directions for Post-SoC/32-bit Networked Sensors

The emergence of low-power 32-bit Systems-on-Chip (SoCs), which integrate a 32-bit MCU, radio, and flash, presents an opportunity to re-examine design points and trade-offs at all levels of the system architecture of networked sensors. To this end, we develop a post-SoC/32-bit design point called Hamilton, showing that using integrated components enables a ~$7 core and shifts hardware modularity to design time. We study the interaction between hardware and embedded operating systems, identifying that (1) post-SoC motes provide lower idle current (5.9 μA) than traditional 16-bit motes, (2) 32-bit MCUs are a major energy consumer (e.g., tick increases idle current >50 times), comparable to radios, and (3) thread-based concurrency is viable, requiring only 8.3 μs of context switch time. We design a system architecture, based on a tickless multithreading operating system, with cooperative/adaptive clocking, advanced sensor abstraction, and preemptive packet processing. Its efficient MCU control improves concurrency with ~30% less energy consumption. Together, these developments set the system architecture for networked sensors in a new direction.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Evaluating In-Clique and Topological Parallelism Strategies for Junction Tree-Based Bayesian Inference Algorithm on the Cray XMT

Long viewed as a strong statistical inference technique, Bayesian networks have emerged to be an important class of applications for high-performance computing. We have applied an architecture-conscious approach to parallelizing the Lauritzen-Spiegelhalter Junction Tree algorithm for exact inferencing in Bayesian networks. In optimizing the Junction Tree algorithm, we have implemented both in-clique and topological parallelism strategies to best leverage the fine-grained synchronization and massive-scale multithreading of the Cray XMT architecture. Two topological techniques were developed to parallelize the evidence propagation process through the Bayesian network. One technique involves performing intelligent scheduling of junction tree nodes based on its topology and relative size. The second technique involves decomposing the junction tree into a much finer tree-like representation to offer much more opportunities for parallelism. We evaluate these optimizations on five different Bayesian networks and report our findings and observations. Another important contribution of this paper is to demonstrate the application of massive-scale multithreading for load balancing and use of implicit parallelism-based compiler optimizations in designing scalable inferencing algorithms.

99 GENERAL AND MISCELLANEOUS↗

iHarmonizer: improving the disk efficiency of I/O-intensive multithreaded codes

Challenged by serious power and thermal constraints, and limited by available instruction-level parallelism, processor designs have evolved to multi-core architectures. These architectures, many augmented with native simultaneous multithreading, are driving software developers to use multithreaded programs to exploit thread-level parallelism. While multithreading is well-known to introduce concerns of data dependency and CPU load balance, less known is that the uncertainty of relative progress of thread execution can cause patterns of I/O requests, issued by different threads, to be effectively random and so significantly degrade hard-disk efficiency. This effect can severely offset the performance gains from parallel execution, especially for I/O-intensive programs. Retaining the benefits of multithreading while not reducing I/O efficiency is an urgent and challenging problem. We propose a user-level scheme, iHarmonizer, to streamline the servicing of I/O requests from multiple threads in OpenMP programs. Specifically, we use the compiler to insert code into OpenMP programs so that data usage can be transmitted at run time to a supporting run-time library; this library in turn prefetches data in a disk-friendly way and coordinates threads execution according to the availability of their requested data. Transparently to the programmer, iHarmonizer makes a multithreaded program I/O efficient while maintaining the benefits of parallelism. Our experiments show that iHarmonizer can significantly speed up the execution of a representative set of I/O-intensive scientific benchmarks.

97 MATHEMATICS AND COMPUTING↗

Multithreaded sparse matrix-matrix multiplication for many-core and GPU architectures

Sparse matrix-matrix multiplication is a key kernel that has applications in several domains such as scientific computing and graph analysis. Several algorithms have been studied in the past for this foundational kernel. In this work, we develop parallel algorithms for sparse matrix-matrix multiplication with a focus on performance portability across different high performance computing architectures. The performance of these algorithms depend on the data structures used in them. We compare different types of accumulators in these algorithms and demonstrate the performance difference between these data structures. Furthermore, we develop a meta-algorithm, kkSpGEMM, to choose the right algorithm and data structure based on the characteristics of the problem. We show performance comparisons on three architectures and demonstrate the need for the community to develop two phase sparse matrix-matrix multiplication implementations for efficient reuse of the data structures involved.

97 MATHEMATICS AND COMPUTING↗

Optimizing the hypre solver for manycore and GPU architectures

The solution of large-scale combustion problems with codes such as Uintah on modern computer architectures requires the use of multithreading and GPUs to achieve performance. Uintah uses a low-Mach number approximation that requires iteratively solving a large system of linear equations. The Hypre iterative solver has solved such systems in a scalable way for Uintah, but the use of OpenMP with Hypre leads to at least slowdown due to OpenMP overheads. The proposed solution uses the MPI Endpoints within Hypre, where each team of threads acts as a different MPI rank. This approach minimizes OpenMP synchronization overhead and performs as fast or (up to 1.44) faster than Hypre's MPI-only version, and allows the rest of Uintah to be optimized using OpenMP. The profiling of the GPU version of Hypre shows the bottleneck to be the launch overhead of thousands of micro-kernels. The GPU performance was improved by fusing these micro-kernels and was further optimized by using Cuda-aware MPI, resulting in an overall speedup of 1.16—1.44 compared to the baseline GPU implementation. The above optimization strategies were published in the International Conference on Computational Science 2020 [1]. This work extends the previously published research by carrying out the second phase of communication-centered optimizations in Hypre to improve its scalability on large-scale supercomputers. Additionally, this includes an efficient non-blocking inter-thread communication scheme, communication-reducing patch assignment, and expression of logical communication parallelism to a new version of the MPICH library that utilizes the underlying network parallelism [2]. The above optimizations avoid communication bottlenecks previously observed during strong scaling and improve performance by up to 2 on 256 nodes of Intel Knight's Landing processor.

97 MATHEMATICS AND COMPUTING↗

Input-independent, Scalable and Fast String Matching on the Cray XMT

String searching is at the core of many security and network applications like search engines, intrusion detection systems, virus scanners and spam filters. The growing size of on-line content and the increasing wire speeds push the need for fast, and often real- time, string searching solutions. For these conditions, many software implementations (if not all) targeting conventional cache-based microprocessors do not perform well. They either exhibit overall low performance or exhibit highly variable performance depending on the types of inputs. For this reason, real-time state of the art solutions rely on the use of either custom hardware or Field-Programmable Gate Arrays (FPGAs) at the expense of overall system flexibility and programmability. This paper presents a software based implementation of the Aho-Corasick string searching algorithm on the Cray XMT multithreaded shared memory machine. Our so- lution relies on the particular features of the XMT architecture and on several algorith- mic strategies: it is fast, scalable and its performance is virtually content-independent. On a 128-processor Cray XMT, it reaches a scanning speed of ≈ 28 Gbps with a performance variability below 10 %. In the 10 Gbps performance range, variability is below 2.5%. By comparison, an Intel dual-socket, 8-core system running at 2.66 GHz achieves a peak performance which varies from 500 Mbps to 10 Gbps depending on the type of input and dictionary size.

97 MATHEMATICS AND COMPUTING↗

Performance Tuning of Fock Matrix and Two-Electron Integral Calculations for NWChem on Leading HPC Platforms

Attaining performance in the evaluation of two-electron repulsion integrals and constructing the Fock matrix is of considerable importance to the computational chemistry community. Due to its numerical complexity improving the performance behavior across a variety of leading supercomputing platforms is an increasing challenge due to the significant diversity in high-performance computing architectures. In this paper, we present our successful tuning methodology for these important numerical methods on the Cray XE6, the Cray XC30, the IBM BG/Q, as well as the Intel Xeon Phi. Our optimization schemes leverage key architectural features including vectorization and simultaneous multithreading, and results in speedups of up to 2.5x compared with the original implementation.

Shan, Hongzhan↗

Center for Programming Models for Scalable Parallel Computing: Future Programming Models

The mission of the pmodel center project is to develop software technology to support scalable parallel programming models for terascale systems. The goal of the specific UD subproject is in the context developing an efficient and robust methodology and tools for HPC programming. More specifically, the focus is on developing new programming models which facilitate programmers in porting their application onto parallel high performance computing systems. During the course of the research in the past 5 years, the landscape of microprocessor chip architecture has witnessed a fundamental change – the emergence of multi-core/many-core chip architecture appear to become the mainstream technology and will have a major impact to for future generation parallel machines. The programming model for shared-address space machines is becoming critical to such multi-core architectures. Our research highlight is the in-depth study of proposed fine-grain parallelism/multithreading support on such future generation multi-core architectures. Our research has demonstrated the significant impact such fine-grain multithreading model can have on the productivity of parallel programming models and their efficient implementation.

97 MATHEMATICS AND COMPUTING↗

Considerations on the Use of Custom Accelerators for Big Data Analytics

Accelerators, including Graphic Processing Units (GPUs) for gen- eral purpose computation, many-core designs with wide vector units (e.g., Intel Phi), have become a common component of many high performance clusters. The appearance of more stable and reliable tools tools that can automatically convert code written in high-level specifications with annotations (such as C or C++) to hardware de- scription languages (High-Level Synthesis - HLS), is also setting the stage for a broader use of reconfigurable devices (e.g., Field Pro- grammable Gate Arrays - FPGAs) in high performance system for the implementation of custom accelerators, helped by the fact that new processors include advanced cache-coherent interconnects for these components. In this chapter, we briefly survey the status of the use of accelerators in high performance systems targeted at big data analytics applications. We argue that, although the progress in the use of accelerators for this class of applications has been sig- nificant, differently from scientific simulations there still are gaps to close. This is particularly true for the ”irregular” behaviors exhibited by no-SQL, graph databases. We focus our attention on the limits of HLS tools for data analytics and graph methods, and discuss a new architectural template that better fits the requirement of this class of applications. We validate the new architectural templates by mod- ifying the Graph Engine for Multithreaded System (GEMS) frame- work to support accelerators generated with such a methodology, and testing with queries coming from the Lehigh University Benchmark (LUBM). The architectural template enables better supporting the task and memory level parallelism present in graph methods by sup- porting a new control model and a enhanced memory interface. We show that out solution allows generating parallel accelerators, pro- viding speed ups with respect to conventional HLS flows. We finally draw conclusions and present a perspective on the use of reconfig- urable devices and Design Automation tools for data analytics.

Data Analytics↗

A Faster Parallel Algorithm and Efficient Multithreaded Implementations for Evaluating Betweenness Centrality on Massive Datasets

We present a new lock-free parallel algorithm for computing betweenness centrality of massive small-world networks. With minor changes to the data structures, our algorithm also achieves better spatial cache locality compared to previous approaches. Betweenness centrality is a key algorithm kernel in the HPCS SSCA#2 Graph Analysis benchmark, which has been extensively used to evaluate the performance of emerging high-performance computing architectures for graph-theoretic computations. We design optimized implementations of betweenness centrality and the SSCA#2 benchmark for two hardware multithreaded systems: a Cray XMT system with the ThreadStorm processor, and a single-socket Sun multicore server with the UltraSparc T2 processor. For a small-world network of 134 million vertices and 1.073 billion edges, the 16-processor XMT system and the 8-core Sun Fire T5120 server achieve TEPS scores (an algorithmic performance count for the SSCA#2 benchmark) of 160 million and 90 million respectively, which corresponds to more than a 2X performance improvement over the previous parallel implementations. To better characterize the performance of these multithreaded systems, we correlate the SSCA#2 performance results with data from the memory-intensive STREAM and RandomAccess benchmarks. Finally, we demonstrate the applicability of our implementation to analyze massive real-world datasets by computing approximate betweenness centrality for a large-scale IMDb movie-actor network.

architecture comparison↗

Preparing MPICH for exascale

The advent of exascale supercomputers heralds a new era of scientific discovery, yet it introduces significant architectural challenges that must be overcome for MPI applications to fully exploit its potential. Among these challenges is the adoption of heterogeneous architectures, particularly the integration of GPUs to accelerate computation. Additionally, the complexity of multithreaded programming models has also become a critical factor in achieving performance at scale. The efficient utilization of hardware acceleration for communication, provided by modern NICs, is also essential for achieving low latency and high throughput communication in such complex systems. In response to these challenges, the MPICH library, a high-performance and widely used Message Passing Interface (MPI) implementation, has undergone significant enhancements. Here, this paper presents four major contributions that prepare MPICH for the exascale transition. First, we describe a lightweight communication stack that leverages the advanced features of modern NICs to maximize hardware acceleration. Second, our work showcases a highly scalable multithreaded communication model that addresses the complexities of concurrent environments. Third, we introduce GPU-aware communication capabilities that optimize data movement in GPU-integrated systems. Finally, we present a new datatype engine aimed at accelerating the use of MPI derived datatypes on GPUs. These improvements in the MPICH library not only address the immediate needs of exascale computing architectures but also set a foundation for exploiting future innovations in high-performance computing. By embracing these new designs and approaches, MPICH-derived libraries from HPE Cray and Intel were able to achieve real exascale performance on OLCF Frontier and ALCF Aurora respectively.

Guo, Yanfei [Argonne National Laboratory (ANL), Ar↗

Parallel Computing in SCALE

The SCALE computational architecture has remained basically the same since its inception 30 years ago, although constituent modules and capabilities have changed significantly. This SCALE concept was intended to provide a framework whereby independent codes can be linked to provide a more comprehensive capability than possible with the individual programs - allowing flexibility to address a wide variety of applications. However, the current system was designed originally for mainframe computers with a single CPU and with significantly less memory than today's personal computers. It has been recognized that the present SCALE computation system could be restructured to take advantage of modern hardware and software capabilities, while retaining many of the modular features of the present system. Preliminary work is being done to define specifications and capabilities for a more advanced computational architecture. This paper describes the state of current SCALE development activities and plans for future development. With the release of SCALE 6.1 in 2010, a new phase of evolutionary development will be available to SCALE users within the TRITON and NEWT modules. The SCALE (Standardized Computer Analyses for Licensing Evaluation) code system developed by Oak Ridge National Laboratory (ORNL) provides a comprehensive and integrated package of codes and nuclear data for a wide range of applications in criticality safety, reactor physics, shielding, isotopic depletion and decay, and sensitivity/uncertainty (S/U) analysis. Over the last three years, since the release of version 5.1 in 2006, several important new codes have been introduced within SCALE, and significant advances applied to existing codes. Many of these new features became available with the release of SCALE 6.0 in early 2009. However, beginning with SCALE 6.1, a first generation of parallel computing is being introduced. In addition to near-term improvements, a plan for longer term SCALE enhancement activities has been developed to provide an integrated framework for future methods development. Some of the major components of the SCALE parallel computing development plan are parallelization and multithreading of computationally intensive modules and redesign of the fundamental SCALE computational architecture.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Contention Modeling for Multithreaded Distributed Shared Memory Machines: The Cray XMT

Distributed Shared Memory (DSM) machines are a wide class of multi-processor computing systems where a large virtually-shared address space is mapped on a network of physically distributed memories. High memory latency and network contention are two of the main factors that limit performance scaling of such architectures. Modern high-performance computing DSM systems have evolved toward exploitation of massive hardware multi-threading and fine-grained memory hashing to tolerate irregular latencies, avoid network hot-spots and enable high scaling. In order to model the performance of such large-scale machines, parallel simulation has been proved to be a promising approach to achieve good accuracy in reasonable times. One of the most critical factors in solving the simulation speed-accuracy trade-off is network modeling. The Cray XMT is a massively multi-threaded supercomputing architecture that belongs to the DSM class, since it implements a globally-shared address space abstraction on top of a physically distributed memory substrate. In this paper, we discuss the development of a contention-aware network model intended to be integrated in a full-system XMT simulator. We start by measuring the effects of network contention in a 128-processor XMT machine and then investigate the trade-off that exists between simulation accuracy and speed, by comparing three network models which operate at different levels of accuracy. The comparison and model validation is performed by executing a string-matching algorithm on the full-system simulator and on the XMT, using three datasets that generate noticeably different contention patterns.

99 GENERAL AND MISCELLANEOUS↗