Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel application”

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

A Programming Model for Massive Data Parallelism with Data Dependencies

Accelerating processors can often be more cost and energy effective for a wide range of data-parallel computing problems than general-purpose processors. For graphics processor units (GPUs), this is particularly the case when program development is aided by environments such as NVIDIA s Compute Unified Device Architecture (CUDA), which dramatically reduces the gap between domain-specific architectures and general purpose programming. Nonetheless, general-purpose GPU (GPGPU) programming remains subject to several restrictions. Most significantly, the separation of host (CPU) and accelerator (GPU) address spaces requires explicit management of GPU memory resources, especially for massive data parallelism that well exceeds the memory capacity of GPUs. One solution to this problem is to transfer data between the GPU and host memories frequently. In this work, we investigate another approach. We run massively data-parallel applications on GPU clusters. We further propose a programming model for massive data parallelism with data dependencies for this scenario. Experience from micro benchmarks and real-world applications shows that our model provides not only ease of programming but also significant performance gains.

43 PARTICLE ACCELERATORS↗

Lightweight and Statistical Techniques for Petascale PetaScale Debugging

This project investigated novel techniques for debugging scientific applications on petascale architectures. In particular, we developed lightweight tools that narrow the problem space when bugs are encountered. We also developed techniques that either limit the number of tasks and the code regions to which a developer must apply a traditional debugger or that apply statistical techniques to provide direct suggestions of the location and type of error. We extend previous work on the Stack Trace Analysis Tool (STAT), that has already demonstrated scalability to over one hundred thousand MPI tasks. We also extended statistical techniques developed to isolate programming errors in widely used sequential or threaded applications in the Cooperative Bug Isolation (CBI) project to large scale parallel applications. Overall, our research substantially improved productivity on petascale platforms through a tool set for debugging that complements existing commercial tools.

97 MATHEMATICS AND COMPUTING↗

An Effective Current Balancing Method for Inverters With Paralleled Silicon Carbide Power Modules

Silicon carbide (SiC) MOSFET has many superior characteristics in high power applications. Paralleling the SiC devices is an effective approach to increase the current capacity of the system, which also leads to current sharing issues. In this article, both transient and steady-state current sharing issues are studied for the high-power grid-tied inverters with paralleled SiC modules with an effective current balancing strategy proposed to address both issues simultaneously. The steady-state imbalance current due to the parameter mismatch is suppressed by the self-inductance of power cables used for paralleling connection. With the elimination of steady-state imbalance current, the transient imbalance current sensing is much simplified and can be realized by using low-cost current sensors. With the sensed imbalance current, a closed-loop control approach is proposed to adjust the modulation reference signal for each module to mitigate the remaining transient imbalance current. Furthermore, the proposed method is straightforward to implement with enhanced compatibility and flexibility compared to existing methods. Experimental studies are performed using inverters with 2 and 4 power modules paralleled in each phase to demonstrate the effectiveness of the proposed method.

42 ENGINEERING↗

Input/Output of ab-initio nuclear structure calculations for improved performance and portability

Many modern scientific applications rely on highly computation intensive calculations. However, most applications do not concentrate as much on the role that input/output operations can play for improved performance and portability. Parallelizing input/output operations of large files can significantly improve the performance of parallel applications where sequential I/O is a bottleneck. A proper choice of I/O library also offers a scope for making input/output operations portable across different architectures. Thus, use of parallel I/O libraries for organizing I/O of large data files offers great scope in improving performance and portability of applications. In particular, sequential I/O has been identified as a bottleneck for the highly scalable MFDn (Many Fermion Dynamics for nuclear structure) code performing ab-initio nuclear structure calculations. We develop interfaces and parallel I/O procedures to use a well-known parallel I/O library in MFDn. As a result, we gain efficient I/O of large datasets along with their portability and ease of use in the down-stream processing. Even situations where the amount of data to be written is not huge, proper use of input/output operations can boost the performance of scientific applications. Application checkpointing offers enormous performance improvement and flexibility by doing a negligible amount of I/O to disk. Checkpointing saves and resumes application state in such a manner that in most cases the application is unaware that there has been an interruption to its execution. This helps in saving large amount of work that has been previously done and continue application execution. This small amount of I/O provides substantial time saving by offering restart/resume capability to applications. The need for checkpointing in optimization code NEWUOA has been identified and checkpoint/restart capability has been implemented in NEWUOA by using simple file I/O.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Simulation of Large-Scale HPC Architectures

The Extreme-scale Simulator (xSim) is a recently developed performance investigation toolkit that permits running high-performance computing (HPC) applications in a controlled environment with millions of concurrent execution threads. It allows observing parallel application performance properties in a simulated extreme-scale HPC system to further assist in HPC hardware and application software co-design on the road toward multi-petascale and exascale computing. This paper presents a newly implemented network model for the xSim performance investigation toolkit that is capable of providing simulation support for a variety of HPC network architectures with the appropriate trade-off between simulation scalability and accuracy. The taken approach focuses on a scalable distributed solution with latency and bandwidth restrictions for the simulated network. Different network architectures, such as star, ring, mesh, torus, twisted torus and tree, as well as hierarchical combinations, such as to simulate network-on-chip and network-on-node, are supported. Network traffic congestion modeling is omitted to gain simulation scalability by reducing simulation accuracy.

99 GENERAL AND MISCELLANEOUS↗

Exploiting Latent I/O Asynchrony in Petascale Science Applications

We present a collection of techniques for exploiting latent I/O asynchrony which can substantially improve performance in data-intensive parallel applications. Latent asynchrony refers to an application’s tolerance for decoupling ancillary operations from its core computation, and is a property of HPC codes not fully explored by current HPC I/O systems. Decoupling operations such as buffering and staging, reorganization, and format conversion in space and in time from core codes can shorten I/O phases, preserving valuable MPP compute cycles. We describe in this paper DataTaps, IOgraphs, and Metabots, three tools which allow HPC developers to implement decoupled I/O operations. Using these tools, asynchrony can be exploited by data generators which overlap computation with communication, and by data consumers that perform data conversion and reorganization out-of-band and on-demand. In the context of a data-intensive fusion simulation, we show that exploiting latent asynchrony through decoupling of operations can provide significant performance benefits.

Computer Science↗

Global‐view coefficients: a data management solution for parallel quantum Monte Carlo applications

Summary Quantum Monte Carlo (QMC) applications perform simulation with respect to an initial state of the quantum mechanical system, which is often captured by using a cubic B‐spline basis. This representation is stored as a read‐only table of coefficients and accesses to the table are generated at random as part of the Monte Carlo simulation. Current QMC applications, such as QWalk and QMCPACK, replicate this table at every process or node, which limits scalability because increasing the number of processors does not enable larger systems to be run. We present a partitioned global address space approach to transparently managing this data using Global Arrays in a manner that allows the memory of multiple nodes to be aggregated. We develop an automated data management system that significantly reduces communication overheads, enabling new capabilities for QMC codes. Experimental results with QWalk and QMCPACK demonstrate the effectiveness of the data management system. Copyright © 2016 John Wiley & Sons, Ltd.

Niu, Qingpeng↗

Efficiency and Scalability of Barrier Synchronization on NoC Based Many-core Architectures

Interconnects based on Networks-on-Chip are an appealing solution to address future microprocessor designs where, very likely, hundreds of cores will be connected on a single chip. A fundamental role in highly parallelized applications running on many-core architectures will be played by barrier primitives used to synchronize the execution of parallel processes. This paper focuses on the analysis of the efficiency and scalability of different barrier implementations in many-core architectures based on NoCs. Several message passing barrier implementations based on four algorithms (all-to-all, master-slave, butterfly and tree) have been implemented and evaluated for a single-chip target architecture composed of a variable number of cores (from 4 to 128) and different network topologies (mesh, torus, ring, clustered-ring and fat-tree). Using a cycle-accurate simulator, we show the scalability of each barrier for every NoC topology, analyzing and comparing theoretical with real behaviors. We observed that some barrier algorithms, when implemented in hardware or software, show a different scaling behavior with respect to those theoretically expected. We evaluate the efficiency of each combination topology-barrier, demonstrating that, in many cases, simple network topologies can be more efficient than complex and highly connected topologies.

97 MATHEMATICS AND COMPUTING↗

ParaOpt: Automated Application Parameterization and Optimization for the Cloud

The variety of instance types available on cloud platforms offers enormous flexibility to match the requirements of applications with available resources. However, selecting the most suitable instance type and configuring an application to optimally execute on that instance type can be complicated and time-consuming. For example, application parallelism flags must match available cores and problem sizes must be tuned to match available memory. As the search space of application configurations can be enormous, we propose an automated approach, called ParaOpt, to automatically explore and tune application configurations on arbitrary cloud instances. ParaOpt supports arbitrary applications, enables use of custom optimization methods, and can be configured with different optimization targets such as runtime and cost. We evaluate ParaOpt by optimizing genomics, molecular dynamics, and machine learning applications with four types of optimizers. We show with as few as 15 parameterized executions of an application, representing between 1.2%-26.7% of the search space, that ParaOpt is able to identify the optimal configuration in 32.7% of experiments and a near-optimal configuration in 83.2% of cases. As a result of using near-optimal configurations, ParaOpt reduces overall execution time by up to 85.8% when compared with using the default configuration.

Wu, Chaofeng↗

File I/O for MPI Applications in Redundant Execution Scenarios

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.

99 GENERAL AND MISCELLANEOUS↗

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↗

Report for the ASC CSSE L2 Milestone (4873) - Demonstration of Local Failure Local Recovery Resilient Programming Model.

Recovery from process loss during the execution of a distributed memory parallel application is presently achieved by restarting the program, typically from a checkpoint file. Future computer system trends indicate that the size of data to checkpoint, the lack of improvement in parallel file system performance and the increase in process failure rates will lead to situations where checkpoint restart becomes infeasible. In this report we describe and prototype the use of a new application level resilient computing model that manages persistent storage of local state for each process such that, if a process fails, recovery can be performed locally without requiring access to a global checkpoint file. LFLR provides application developers with an ability to recover locally and continue application execution when a process is lost. This report discusses what features are required from the hardware, OS and runtime layers, and what approaches application developers might use in the design of future codes, including a demonstration of LFLR-enabled MiniFE code from the Matenvo mini-application suite.

Heroux, Michael Allen↗

Lightweight and Statistical Techniques for Petascale Debugging: Correctness on Petascale Systems (CoPS) Preliminry Report

Petascale platforms with O(10{sup 5}) and O(10{sup 6}) processing cores are driving advancements in a wide range of scientific disciplines. These large systems create unprecedented application development challenges. Scalable correctness tools are critical to shorten the time-to-solution on these systems. Currently, many DOE application developers use primitive manual debugging based on printf or traditional debuggers such as TotalView or DDT. This paradigm breaks down beyond a few thousand cores, yet bugs often arise above that scale. Programmers must reproduce problems in smaller runs to analyze them with traditional tools, or else perform repeated runs at scale using only primitive techniques. Even when traditional tools run at scale, the approach wastes substantial effort and computation cycles. Continued scientific progress demands new paradigms for debugging large-scale applications. The Correctness on Petascale Systems (CoPS) project is developing a revolutionary debugging scheme that will reduce the debugging problem to a scale that human developers can comprehend. The scheme can provide precise diagnoses of the root causes of failure, including suggestions of the location and the type of errors down to the level of code regions or even a single execution point. Our fundamentally new strategy combines and expands three relatively new complementary debugging approaches. The Stack Trace Analysis Tool (STAT), a 2011 R&D 100 Award Winner, identifies behavior equivalence classes in MPI jobs and highlights behavior when elements of the class demonstrate divergent behavior, often the first indicator of an error. The Cooperative Bug Isolation (CBI) project has developed statistical techniques for isolating programming errors in widely deployed code that we will adapt to large-scale parallel applications. Finally, we are developing a new approach to parallelizing expensive correctness analyses, such as analysis of memory usage in the Memgrind tool. In the first two years of the project, we have successfully extended STAT to determine the relative progress of different MPI processes. We have shown that the STAT, which is now included in the debugging tools distributed by Cray with their large-scale systems, substantially reduces the scale at which traditional debugging techniques are applied. We have extended CBI to large-scale systems and developed new compiler based analyses that reduce its instrumentation overhead. Our results demonstrate that CBI can identify the source of errors in large-scale applications. Finally, we have developed MPIecho, a new technique that will reduce the time required to perform key correctness analyses, such as the detection of writes to unallocated memory. Overall, our research results are the foundations for new debugging paradigms that will improve application scientist productivity by reducing the time to determine which package or module contains the root cause of a problem that arises at all scales of our high end systems. While we have made substantial progress in the first two years of CoPS research, significant work remains. While STAT provides scalable debugging assistance for incorrect application runs, we could apply its techniques to assertions in order to observe deviations from expected behavior. Further, we must continue to refine STAT's techniques to represent behavioral equivalence classes efficiently as we expect systems with millions of threads in the next year. We are exploring new CBI techniques that can assess the likelihood that execution deviations from past behavior are the source of erroneous execution. Finally, we must develop usable correctness analyses that apply the MPIecho parallelization strategy in order to locate coding errors. We expect to make substantial progress on these directions in the next year but anticipate that significant work will remain to provide usable, scalable debugging paradigms.

97 MATHEMATICS AND COMPUTING↗

Benchmarking the effects of operating system interference on extreme-scale parallel machines.

We investigate operating system noise, which we identify as one of the main reasons for a lack of synchronicity in parallel applications. Using a microbenchmark, we measure the noise on several contemporary platforms and find that, even with a general-purpose operating system, noise can be limited if certain precautions are taken. We then inject artificially generated noise into a massively parallel system and measure its influence on the performance of collective operations. Our experiments indicate that on extreme-scale platforms, the performance is correlated with the largest interruption to the application, even if the probability of such an interruption on a single process is extremely small. We demonstrate that synchronizing the noise can significantly reduce its negative influence.

97 MATHEMATICS AND COMPUTING↗

Characterizing Computation-Communication Overlap in Message-Passing Systems

Effective overlap of computation and communication is a well understood technique for latency hiding and can yield significant performance gains for applications on high-end computers. In this report, we describe an instrumentation framework developed for messagepassing systems to characterize the degree of overlap of communication with computation in the execution of parallel applications. The inability to obtain precise time-stamps for pertinent communication events is a significant problem, and is addressed by generation of minimum and maximum bounds on achieved overlap. The overlap measures can aid application developers and system designers in investigating scalability issues. The approach has been used to instrument two MPI implementations as well as the ARMCI system. The implementation resides entirely within the communication library and thus integrates well with existing approaches that operate outside the library. The utility of the framework is demonstrated by analyzing communication-computation overlap for micro-benchmarks and the NAS benchmarks, and the insights obtained are used to modify the NAS SP benchmark, resulting in improved overlap.

97 MATHEMATICS AND COMPUTING↗

Compiler-Directed File Layout Optimization for Hierarchical Storage Systems

File layout of array data is a critical factor that effects the behavior of storage caches, and has so far taken not much attention in the context of hierarchical storage systems. The main contribution of this paper is a compiler-driven file layout optimization scheme for hierarchical storage caches. This approach, fully automated within an optimizing compiler, analyzes a multi-threaded application code and determines a file layout for each disk-resident array referenced by the code, such that the performance of the target storage cache hierarchy is maximized. We tested our approach using 16 I/O intensive application programs and compared its performance against two previously proposed approaches under different cache space management schemes. Our experimental results show that the proposed approach improves the execution time of these parallel applications by 23.7% on average.

Ding, Wei↗