Search NASA⌕ Search

SEARCH · Search NASA

Results for “asynchronous processing”

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 91 records · Page 5

A Task Based Approach for Co-Scheduling Ensemble Workloads on Heterogeneous Nodes

Scientific workflows consist of multiple, connected applications, with data and results flowing from one to another in a pipeline. Traditionally, such workflows are executed in sequential order, storing intermediate data in storage disks. Co-scheduling application workflows concurrently on the same compute nodes would greatly reduce the cost of moving data to/from storage and allow real-time analysis of intermediate results. Nevertheless, most parallel programming runtimes do not allow seamless integration of various applications in a scientific workflow, in part due to the complexity of managing data and resources. The situation is even more complicated for heterogeneous systems. In this work we extend the Minos Computing Library (MCL) runtime to accelerate pipe-lined and parallel workloads where multiple applications are running in the same system. MCL’s asynchronous task library and runtime dynamically manages resources to allow co-scheduling of multiple processes sharing heterogeneous resources. In addition, we design a custom ex- tension of the Open Compute Language (OpenCL) to enable multiple processes to share device memory. We enable MCL to coordinate these shared buffers to allow for easy, fast data sharing between applications. Using malleable micro-benchmarks and two application workflows that combine scientific simulation and AI-based analysis, we show that our method outperforms traditional approaches.

Index Terms—Parallel systems, Scheduling and Task ↗

Parallel Asynchronous Distributed-Memory Maximal Independent Set Algorithm with Work Ordering

The maximal independent set (MIS) graph problem arises in many applications such as computer vision, information theory, molecular biology, and process scheduling. The growing scale of graph data suggests the use of distributed memory hardware as a cost-effective approach to providing necessary compute and memory resources. Existing distributed memory parallel MIS algorithms rely on synchronous communication and use techniques such as subgraph computations. In this paper, we present an asynchronous distributed-memory parallel graph algorithm that relies on a virtual directed acyclic graph (DAG) that is created during the algorithm execution. We introduce two additional algorithms that save computations by ordering generated work. The first algorithm applies ordering globally to reduce computations, and the second algorithm applies ordering locally at the level of threads to minimize the synchronization overhead. We use two different implementations of Luby’s algorithm variants as baseline to compare the performance of the presented algorithms: (1) vertex-centric Luby A and Luby B implementations, and (2) the CombBLAS linear-algebra Luby A implementation. Results show that proposed algorithms outperform both implementations of Luby algorithms, especially in distributed execution. Furthermore, we show that for low-diameter graphs the algorithm that applies global ordering scales better than other algorithms and for high diameter graphs the original asynchronous algorithm and thread-level ordering algorithm show better performance.

maximal independent set (MIS), distributed memory ↗

SWAP-Assembler: scalable and efficient genome assembly towards thousands of cores

Background: There is a widening gap between the throughput of massive parallel sequencing machines and the ability to analyze these sequencing data. Traditional assembly methods requiring long execution time and large amount of memory on a single workstation limit their use on these massive data. Results: This paper presents a highly scalable assembler named as SWAP-Assembler for processing massive sequencing data using thousands of cores, where SWAP is an acronym for Small World Asynchronous Parallel model. In the paper, a mathematical description of multi-step bi-directed graph (MSG) is provided to resolve the computational interdependence on merging edges, and a highly scalable computational framework for SWAP is developed to automatically preform the parallel computation of all operations. Graph cleaning and contig extension are also included for generating contigs with high quality. Experimental results show that SWAP-Assembler scales up to 2048 cores on Yanhuang dataset using only 26 minutes, which is better than several other parallel assemblers, such as ABySS, Ray, and PASHA. Results also show that SWAP-Assembler can generate high quality contigs with good N50 size and low error rate, especially it generated the longest N50 contig sizes for Fish and Yanhuang datasets. Conclusions: In this paper, we presented a highly scalable and efficient genome assembly software, SWAPAssembler. Compared with several other assemblers, it showed very good performance in terms of scalability and contig quality. This software is available at: https://sourceforge.net/projects/swapassembler

59 BASIC BIOLOGICAL SCIENCES↗

Architecture and method for a burst buffer using flash technology

A parallel supercomputing cluster includes compute nodes interconnected in a mesh of data links for executing an MPI job, and solid-state storage nodes each linked to a respective group of the compute nodes for receiving checkpoint data from the respective compute nodes, and magnetic disk storage linked to each of the solid-state storage nodes for asynchronous migration of the checkpoint data from the solid-state storage nodes to the magnetic disk storage. Each solid-state storage node presents a file system interface to the MPI job, and multiple MPI processes of the MPI job write the checkpoint data to a shared file in the solid-state storage in a strided fashion, and the solid-state storage node asynchronously migrates the checkpoint data from the shared file in the solid-state storage to the magnetic disk storage and writes the checkpoint data to the magnetic disk storage in a sequential fashion.

97 MATHEMATICS AND COMPUTING↗

Distributed Automatic Differentiation for Ptychography

Synchrotron radiation light source facilities are leading the way to ultrahigh resolution X-ray imaging. High resolution imaging is essential to understanding the fundamental structure and interaction of materials at the smallest length scale possible. Diffraction based methods achieve nanoscale imaging by replacing traditional objective lenses by pixelated area detectors and computational image reconstruction. Among these methods, ptychography is quickly becoming the standard for sub-30 nanometer imaging of extended samples, but at the expense of increasingly high data rates and volumes. This paper presents a new distributed algorithm for solving the ptychographic image reconstruction problem based on automatic differentiation. Input datasets are subdivided between multiple graphics processing units (GPUs); each subset of the problem is then solved either entirely independent of other subsets (asynchronously) or through sharing gradient information with other GPUs (synchronously). The algorithm was evaluated on simulated and real data acquired at the Advanced Photon Source, scaling up to 192 GPUs. The synchronous variant of our method outperformed an existing multi-GPU implementation in terms of accuracy while running at a comparable execution time.

distributed algorithms↗

UMap: An application-oriented user level memory mapping library

Exploiting the prominent role of complex memories in exascale node architecture, the UMap page fault handler offers new capabilities to access large memory-mapped data sets directly. UMap provides flexible configuration options to customize page handling to each application, including analysis of massive observational and simulation data sets. The high-performance design features I/O decoupling, dynamic load balancing, and application-level controls. Page faults triggered by application threads and processes accessing data mapped to a UMapp’ed region are handled via the Linux userfaultfd protocol, an asynchronous message-oriented kernel-user communication mechanism that avoids the context switch penalty of traditional signal fault handlers. UMap is fully open source. In this paper, we give an overview of the UMap library architecture, its extensible plugin architecture, and the use/performance of UMap in emerging heterogeneous memory hierarchies such as near-node Non-volatile Memory (NVM) and network attached memories. We highlight new capabilities in two pagefault management plugins, the NetworkStore and SparseStore. We demonstrate the integration between UMap and multiple ECP products including Caliper, Metall, ZFP, Mochi, and Ripples.

97 MATHEMATICS AND COMPUTING↗

High-frequency Data Integration for Landscape Model Calibration of Carbon Fluxes Across Diverse Tidal Marshes

Terrestrial Aquatic Interfaces (TAIs), and tidal wetlands in particular, store large amounts of carbon yet are not well represented in Earth System Models (ESMs). Predictions of carbon cycling and greenhouse gas (GHG) emissions in tidal wetlands are highly uncertain. Eddy covariance (EC) towers provide ecosystem-scale GHG flux data at a temporal resolution (every 30min) that is helpful for parameterizing and improving mechanistic realism in ESMs. We propose to use a network of eddy covariance towers and standardized ancillary data streams, along with mesocosm experiments and statistical analyses, across diverse tidal wetlands of North America to develop and improve biogeochemical modeling at the TAI. Our overarching objective is to improve understanding and process-based modeling of gross primary productivity (GPP) and CH4 emission responses, both non-linear and asynchronous, to stressors including plant inundation, disturbance, salinity and nitrogen loading.

54 ENVIRONMENTAL SCIENCES↗

Ultrascalable petaflop parallel supercomputer

A massively parallel supercomputer of petaOPS-scale includes node architectures based upon System-On-a-Chip technology, where each processing node comprises a single Application Specific Integrated Circuit (ASIC) having up to four processing elements. The ASIC nodes are interconnected by multiple independent networks that optimally maximize the throughput of packet communications between nodes with minimal latency. The multiple networks may include three high-speed networks for parallel algorithm message passing including a Torus, collective network, and a Global Asynchronous network that provides global barrier and notification functions. These multiple independent networks may be collaboratively or independently utilized according to the needs or phases of an algorithm for optimizing algorithm processing performance. The use of a DMA engine is provided to facilitate message passing among the nodes without the expenditure of processing resources at the node.

97 MATHEMATICS AND COMPUTING↗

Polarized Substructural Session Types

Concurrent processes can be extremely difficult to reason about, both for programmers and formally. One approach to coping with this difficulty is to study new programming languages and type features such as Session Types. Session types take as their conceptual notion of concurrency as a collection of processes linked together via channels and provide type-level coordination between processes using these channels. Logically motivated programming languages exploit the idea that providing a proof of a theorem in a logic is similar to proving that a given term has a particular type in a programming language and vice versa. These connections can be interesting for a few different reasons. First, when language and logic are independently discovered and independently useful, the existence of a connection suggests that both are onto some fundamentally important idea. Additionally, a connection provides a basis both for sanity checking our ideas and also can be fruitful grounds for inspiration by seeing how variants of either the logic or the language are reflected through the connection. This thesis primarily describes an exploration of logically motivated session types, SILL. Polarization, classifying propositions as either positive or negative, provides a natural way to describe a logically based session typing language with asynchronous communication while retaining a semantics that is reasonably implementable. Additionally, polarization gives us a way to smoothly integrate synchronous channels into SILL without needing a semantic extension. When combined with Adjoint Logic, this gives us an ability to incorporate a variety of modalities with relatively little work. From a practical perspective, this gives SILL access to persistent processes and garbage collection. We additionally explore a trio of loosely related extensions to SILL, and their logical connections, inspired by the above results: bundled message passing to reduce the number of communications performed by processes; racy programs, enabled by a select/epoll-like mechanism; and asynchronous receiving, an almost generalization of the basic asynchronous semantics. We have three different implementations of SILL: a simple but relatively full featured interpreter written in OCaml; a fragment of SILL as an embedded domain specific language in Haskell; and a cleaner version of the same in Idris. Lastly, we show that Liquid Types and Session Types are compatible. This gives us one notion of a dependently session typed language.

97 MATHEMATICS AND COMPUTING↗

Structured Adaptive Mesh Refinement Adaptations to Retain Performance Portability With Increasing Heterogeneity

Adaptive mesh refinement (AMR) is an important method that enables many mesh-based applications to run at effectively higher resolution within limited computing resources by allowing high resolution only where really needed. This advantage comes at a cost, however: greater complexity in the mesh management machinery and challenges with load distribution. With the current trend of increasing heterogeneity in hardware architecture, AMR presents an orthogonal axis of complexity. Additionally, the usual techniques, such as asynchronous communication and hierarchy management for parallelism and memory that are necessary to obtain reasonable performance are very challenging to reason about with AMR. Different groups working with AMR are bringing different approaches to this challenge. Here, we examine the design choices of several AMR codes and also the degree to which demands placed on them by their users influence these choices.

42 ENGINEERING↗

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗

Power Minimization techniques for Networked Data Centers.

Our objective is to develop a mathematical model to optimize energy consumption at multiple levels in networked data centers, and develop abstract algorithms to optimize not only individual servers, but also coordinate the energy consumption of clusters of servers within a data center and across geographically distributed data centers to minimize the overall energy cost and consumption of brown energy of an enterprise. In this project, we have formulated a variety of optimization models, some stochastic others deterministic, and have obtained a variety of qualitative results on the structural properties, robustness, and scalability of the optimal policies. We have also systematically derived from these models decentralized algorithms to optimize energy efficiency, analyzed their optimality and stability properties. Finally, we have conducted preliminary numerical simulations to illustrate the behavior of these algorithms. We draw the following conclusion. First, there is a substantial opportunity to minimize both the amount and the cost of electricity consumption in a network of datacenters, by exploiting the fact that traffic load, electricity cost, and availability of renewable generation fluctuate over time and across geographical locations. Judiciously matching these stochastic processes can optimize the tradeoff between brown energy consumption, electricity cost, and response time. Second, given the stochastic nature of these three processes, real-time dynamic feedback should form the core of any optimization strategy. The key is to develop decentralized algorithms that can be implemented at different parts of the network as simple, local algorithms that coordinate through asynchronous message passing.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

HYPPO v1.0.0

HYPPO is a hyperparameter optimization (HPO) software that uses surrogate modeling and uncertainty quantification to provide reliable neural network architectures. Similar state-of-the-art HPO technologies do not use such features and mostly perform random searches to explore the parameter space, which can easily become computationally expensive. Using HYPPO, the user is able to reduce by an order of magnitude the number of evaluations necessary to identify the most optimal region in the hyperparameter space. Finally, using asynchronous nested parallelism, we are able to decrease by two orders of magnitude the amount of time required to complete the HPO process.

Dumont, Vincent↗

Mapping Dense LU Factorization on Multicore Supercomputer Nodes

Dense LU factorization is a prominent benchmark used to rank the performance of supercomputers. Many implementations use block-cyclic distributions of matrix blocks onto a two-dimensional process grid. The process grid dimensions drive a trade-off between communication and computation and are architecture- and implementation-sensitive. The critical panel factorization steps can be made less communication-bound by overlapping asynchronous collectives for pivoting with the computation of rank-k updates. By shifting the computation-communication trade-off, a modified block-cyclic distribution can beneficially exploit more available parallelism on the critical path, and reduce panel factorization's memory hierarchy contention on now-ubiquitous multicore architectures. During active panel factorization, rank-1 updates stream through memory with minimal reuse. In a column-major process grid, the performance of this access pattern degrades as too many streaming processors contend for access to memory. A block-cyclic mapping in the row-major order does not encounter this problem, but consequently sacrifices node and network locality in the critical pivoting steps. We introduce 'striding' to vary between the two extremes of row- and column-major process grids. The maximum available parallelism in the critical path work (active panel factorization, triangular solves, and subsequent broadcasts) is bounded by the length or width of the process grid. Increasing one dimension of the process grid decreases the number of distinct processes and nodes in the other dimension. To increase the harnessed parallelism in both dimensions, we start with a tall process grid. We then apply periodic 'rotation' to this grid to restore exploited parallelism along the row to previous levels. As a test-bed for further mapping experiments, we describe a dense LU implementation that allows a block distribution to be defined as a general function of block to processor. Other mappings can be tested wit...

Computer Science↗

Multi-petascale highly efficient parallel supercomputer

A Multi-Petascale Highly Efficient Parallel Supercomputer of 100 petaflop-scale includes node architectures based upon System-On-a-Chip technology, where each processing node comprises a single Application Specific Integrated Circuit (ASIC). The ASIC nodes are interconnected by a five dimensional torus network that optimally maximize the throughput of packet communications between nodes and minimize latency. The network implements collective network and a global asynchronous network that provides global barrier and notification functions. Integrated in the node design include a list-based prefetcher. The memory system implements transaction memory, thread level speculation, and multiversioning cache that improves soft error rate at the same time and supports DMA functionality allowing for parallel processing message-passing.

97 MATHEMATICS AND COMPUTING↗

Network-friendly one-sided communication through multinode cooperation on petascale cray xt5 systems

ne-sided communication is important to enable asynchronous communication and data movement for Global Address Space (GAS) programming models. Such communication is typically realized through direct messages between initiator and target processes. For petascale systems with 10,000s of nodes and 100,000s of cores, these direct messages require dedicated communication buffers and/or channels, which can lead to significant scalability challenges for GAS programming models. In this paper, we describe a network-friendly communication model, multinode cooperation, to enable indirect one-sided communication. Compute nodes work together to handle one-sided requests through (1) request forwarding in which one node can intercept a request and forward it to a target node, and (2) request aggregation in which one node can aggregate many requests to a target node. We have implemented multinode cooperation for a popular GAS runtime library, Aggregate Remote Memory Copy Interface (ARMCI). Our experimental results on a large-scale Cray XT5 system demonstrate that, multinode cooperation is able to greatly increase the memory scalability by reducing the number of communication buffers. In addition, multinode cooperation improves the resiliency of GAS runtime system to network contention. Furthermore, multinode cooperation can benefit the performance of scientific applications. In one case, it reduces the total execution time of an NWChem application by 52%.

99 GENERAL AND MISCELLANEOUS↗

Network-friendly one-sided communication through multinode cooperation on petascale cray xt5 systems

ne-sided communication is important to enable asynchronous communication and data movement for Global Address Space (GAS) programming models. Such communication is typically realized through direct messages between initiator and target processes. For petascale systems with 10,000s of nodes and 100,000s of cores, these direct messages require dedicated communication buffers and/or channels, which can lead to significant scalability challenges for GAS programming models. In this paper, we describe a network-friendly communication model, multinode cooperation, to enable indirect one-sided communication. Compute nodes work together to handle one-sided requests through (1) request forwarding in which one node can intercept a request and forward it to a target node, and (2) request aggregation in which one node can aggregate many requests to a target node. We have implemented multinode cooperation for a popular GAS runtime library, Aggregate Remote Memory Copy Interface (ARMCI). Our experimental results on a large-scale Cray XT5 system demonstrate that, multinode cooperation is able to greatly increase the memory scalability by reducing the number of communication buffers. In addition, multinode cooperation improves the resiliency of GAS runtime system to network contention. Furthermore, multinode cooperation can benefit the performance of scientific applications. In one case, it reduces the total execution time of an NWChem application by 52%.

Tipparaju, Vinod↗

Dense LU Factorization on Multicore Supercomputer Nodes

Dense LU factorization is a prominent benchmark used to rank the performance of supercomputers. Many implementations, including the reference code HPL, use block-cyclic distributions of matrix blocks onto a two-dimensional process grid. The process grid dimensions drive a trade-off between communication and computation and are architecture- and implementation-sensitive. We show how the critical panel factorization steps can be made less communication-bound by overlapping asynchronous collectives for pivot identification and exchange with the computation of rank-k updates. By shifting this trade-off, a modified block-cyclic distribution can beneficially exploit more available parallelism on the critical path, and reduce panel factorization's memory hierarchy contention on now-ubiquitous multi-core architectures. The missed parallelism in traditional block-cyclic distributions arises because active panel factorization, triangular solves, and subsequent broadcasts are spread over single process columns or rows (respectively) of the process grid. Increasing one dimension of the process grid decreases the number of distinct processes in the other dimension. To increase parallelism in both dimensions, periodic 'rotation' is applied to the process grid to recover the row-parallelism lost by a tall process grid. During active panel factorization, rank-1 updates stream through memory with minimal reuse. In a column-major process grid, the performance of this access pattern degrades as too many streaming processors contend for access to memory. A block-cyclic mapping in the more popular row-major order does not encounter this problem, but consequently sacrifices node and network locality in the critical pivoting steps. We introduce 'striding' to vary between the two extremes of row- and column-major process grids. As a test-bed for further mapping experiments, we describe a dense LU implementation that allows a block distribution to be defined as a general function of block to processor. Other mappings can be tested with only small, local changes to the code.

99 GENERAL AND MISCELLANEOUS↗