Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “asynchronous”

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

Evaluating asynchronous Schwarz solvers on GPUs

With the commencement of the exascale computing era, we realize that the majority of the leadership supercomputers are heterogeneous and massively parallel. Even a single node can contain multiple co-processors such as GPUs and multiple CPU cores. For example, ORNL’s Summit accumulates six NVIDIA Tesla V100 GPUs and 42 IBM Power9 cores on each node. Synchronizing across compute resources of multiple nodes can be prohibitively expensive. Hence, it is necessary to develop and study asynchronous algorithms that circumvent this issue of bulk-synchronous computing. In this study, we examine the asynchronous version of the abstract Restricted Additive Schwarz method as a solver. We do not explicitly synchronize, but allow the communication between the sub-domains to be completely asynchronous, thereby removing the bulk synchronous nature of the algorithm. We accomplish this by using the one-sided Remote Memory Access (RMA) functions of the MPI standard. We study the benefits of using such an asynchronous solver over its synchronous counterpart. We also study the communication patterns governed by the partitioning and the overlap between the sub-domains on the global solver. Finally, we show that this concept can render attractive performance benefits over the synchronous counterparts even for a well-balanced problem.

Nayak, Pratik↗

Asynchronous Iterative Solvers for Extreme-Scale Computing

The Asynchronous Iterative Solvers for Extreme-Scale Computing (AsyncIS) project aims to explore more efficient numerical algorithms by decreasing their overhead. AsyncIS does this by replacing the outer Krylov subspace solver with an asynchronous optimized Schwarz method, thereby removing the global synchronization and bulk synchronous operations typically used in numerical codes. AsyncIS—a U.S. Department of Energy (DOE)-funded collaboration between Georgia Tech, the University of Tennessee, Knoxville, Temple University, and Sandia National Laboratories—also focuses on the development and optimization of asynchronous preconditioners (i.e., preconditioners that are generated and/or applied in an asynchronous fashion). The novel preconditioning algorithms that provide fine-grained parallelism enable preconditioned Krylov solvers to run efficiently on large-scale distributed systems and manycore accelerators like GPUs.

97 MATHEMATICS AND COMPUTING↗

Control circuits for asynchronous circuits

The described embodiments include a computing device with one or more asynchronous circuits and control circuits that control the operation of the asynchronous circuits. In some embodiments, the control circuits are arranged in a hierarchy with a top-level control circuit atop the hierarchy and one or more local control circuits lower in the hierarchy. In these embodiments, the top-level control circuit processes operating information for the one or more asynchronous circuits and/or other functional blocks in the computing device to determine an operating state for the computing device. Based on the operating state, the top-level control circuit communicates commands to the local control circuits to cause the local control circuits to operate in corresponding operating modes. Based on a corresponding operating mode command, each local control circuit sets one or more operating parameters for corresponding asynchronous circuits (and/or one or more other functional blocks).

42 ENGINEERING↗

Conflux – An Asynchronous 2-to-1 Multiplexor for Time-Division Multiplexing and Clock-less, Token-less Readout

This paper describes an independent, self-arbitrating asynchronous two-to-one multiplexor called Conflux. Conflux can be used to implement clock-less and token-less Time-Division Multiplexing between two sources. This capability can then be used to create a clock-less and token-less aggregate data bus which can be used as a complete asynchronous readout architecture for a chip, a part of a chip or an entire multi-chip system. Conflux utilizes a classic, four-phased asynchronous handshake on both of its input ports as well as on its output port. Asynchronous Request is bundled with Data to ensure consistent propagation delay. The Request is also differential to account for propagation delay differences between Logical Ones and Zeros. Arbitration between the two Conflux input streams is accomplished by three Set-Reset latches. Finally, Conflux cells were used successfully to implement the readout architecture of the FCP130 prototype chip. Tests indicate single Conflux stage delays of 1.8ns and a bandwidth of approximately 20 Gbps in 130nm CMOS.

43 PARTICLE ACCELERATORS↗

Scalable Asynchronous Domain Decomposition Solvers

We discuss how parallel implementations of linear iterative solvers generally alternate between phases of data exchange and phases of local computation. Increasingly large problem sizes and more heterogeneous compute architectures make load balancing and the design of low latency network interconnects that are able to satisfy the communication requirements of linear solvers very challenging tasks. In particular, global communication patterns such as inner products become increasingly limiting at scale. We explore the use of asynchronous communication based on one-sided Message Passing Interface primitives in the context of domain decomposition solvers. In particular, a scalable asynchronous two-level Schwarz method is presented. We discuss practical issues encountered in the development of a scalable solver and show experimental results obtained on a state-of-the-art supercomputer system that illustrate the benefits of asynchronous solvers in load balanced as well as load imbalanced scenarios. Using the novel method, we can observe speedups of up to four times over its classical synchronous equivalent.

97 MATHEMATICS AND COMPUTING↗

Asynchronous Programming in UPC: A Case Study and Potential for Improvement

In a traditional Partitioned Global Address Space language like UPC, an application programmer works with the model of a static set of threads performing locality-aware accesses on a global address space. On the other hand, asynchronous programming provides a simple interface for expressing the concurrency in dynamic, irregular algorithms, with the prospect of efficient portable execution from sophisticated runtime schemes handling the exposed concurrency. In this paper, we adopt the asynchronous style of programming to parallelize a nested, tree-based code in UPC. To maximize performance without losing the ease of application programming, we design Asynchronous Remote Methods as a potential extension to the UPC standard. Our prototype implementation of this construct in Berkeley UPC yields within 7% of ideal performance and 20-fold improvement over the original Standard UPC solution in some cases.

97 MATHEMATICS AND COMPUTING↗

aphBO-2GP-3B: a budgeted asynchronous parallel multi-acquisition functions for constrained Bayesian optimization on high-performing computing architecture

High-fidelity complex engineering simulations are often predictive, but also computationally expensive and often require substantial computational efforts. The mitigation of computational burden is usually enabled through parallelism in high-performance cluster (HPC) architecture. Optimization problems associated with these applications is a challenging problem due to the high computational cost of the high-fidelity simulations. In this paper, an asynchronous parallel constrained Bayesian optimization method is proposed to efficiently solve the computationally expensive simulation-based optimization problems on the HPC platform, with a budgeted computational resource, where the maximum number of simulations is a constant. The advantage of this method are three-fold. Firstly, the efficiency of the Bayesian optimization is improved, where multiple input locations are evaluated parallel in an asynchronous manner to accelerate the optimization convergence with respect to physical runtime. This efficiency feature is further improved so that when each of the inputs is finished, another input is queried without waiting for the whole batch to complete. Second, the proposed method can handle both known and unknown constraints. Third, the proposed method samples several acquisition functions based on their rewards using a modified GP-Hedge scheme. The proposed framework is termed aphBO-2GP-3B, which means asynchronous parallel hedge Bayesian optimization with two Gaussian processes and three batches. The numerical performance of the proposed framework aphBO-2GP-3B is comprehensively benchmarked using 16 numerical examples, compared against other 6 parallel Bayesian optimization variants and 1 parallel Monte Carlo as a baseline, and demonstrated using two real-world high-fidelity expensive industrial applications. The first engineering application is based on finite element analysis (FEA) and the second one is based on computational fluid dynamics (CFD) simulations.

97 MATHEMATICS AND COMPUTING↗

Frog: Asynchronous Graph Processing on GPU with Hybrid Coloring Model

GPUs have been increasingly used to accelerate graph processing for complicated computational problems regarding graph theory. Many parallel graph algorithms adopt the asynchronous computing model to accelerate the iterative convergence. Unfortunately, the consistent asynchronous computing requires locking or atomic operations, leading to significant penalties/overheads when implemented on GPUs. As such, coloring algorithm is adopted to separate the vertices with potential updating conflicts, guaranteeing the consistency/correctness of the parallel processing. Common coloring algorithms, however, may suffer from low parallelism because of a large number of colors generally required for processing a large-scale graph with billions of vertices. We propose a light-weight asynchronous processing framework called Frog with a preprocessing/hybrid coloring model. The fundamental idea is based on Pareto principle (or 80-20 rule) about coloring algorithms as we observed through masses of realworld graph coloring cases. We find that a majority of vertices (about 80%) are colored with only a few colors, such that they can be read and updated in a very high degree of parallelism without violating the sequential consistency. Accordingly, our solution separates the processing of the vertices based on the distribution of colors. In this work, we mainly answer three questions: (1) how to partition the vertices in a sparse graph with maximized parallelism, (2) how to process large-scale graphs that cannot fit into GPU memory, and (3) how to reduce the overhead of data transfers on PCIe while processing each partition. Here, we conduct experiments on real-world data (Amazon, DBLP, YouTube, RoadNet-CA, WikiTalk and Twitter) to evaluate our approach and make comparisons with well-known non-preprocessed (such as Totem, Medusa, MapGraph and Gunrock) and preprocessed (Cusha) approaches, by testing four classical algorithms (BFS, PageRank, SSSP and CC). On all the tested applications and datasets, Frog is able to significantly outperform existing GPU-based graph processing systems except Gunrock and MapGraph. MapGraph gets better performance than Frog when running BFS on RoadNet-CA. The comparison between Gunrock and Frog is inconclusive. Frog can outperform Gunrock more than 1.04X when running PageRank and SSSP, while the advantage of Frog is not obvious when running BFS and CC on some datasets especially for RoadNet-CA.

97 MATHEMATICS AND COMPUTING↗

ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing

The next generation HPC and data centers are likely to be reconfigurable and data-centric due to the trend of hardware specialization and the emergence of data-driven applications. In this work, we propose ARENA – an asynchronous reconfigurable accelerator ring architecture as a potential scenario on how the future HPC and data centers will be like. Despite using the coarse-grained reconfigurable arrays (CGRAs) as the substrate platform, our key contribution is not only the CGRA-cluster design itself, but also the ensemble of a new architecture and programming model that enables asynchronous tasking across a cluster of reconfigurable nodes, so as to bring specialized computation to the data rather than the reverse. We presume distributed data storage without asserting any prior knowledge on the data distribution. Hardware specialization occurs at runtime when a task finds the majority of data it requires are available at the present node. In other words, we dynamically generate specialized CGRA accelerators where the data reside. The asynchronous tasking for bringing computation to data is achieved by circulating the task token, which describes the dataflow graphs to be executed for a task, among the CGRA cluster connected by a fast ring network. Evaluations on a set of HPC and data-driven applications across different domains show that ARENA can provide better parallel scalability with reduced data movement (53.9 percent). Compared with contemporary compute-centric parallel models, ARENA can bring on average 4.37× speedup. The synthesized CGRAs and their task-dispatchers only occupy 2.93mm 2 chip area under 45nm process technology and can run at 800MHz with on average 759.8mW power consumption. ARENA also supports the concurrent execution of multi-applications, offering ideal architectural support for future high-performance parallel computing and data analytics systems.

97 MATHEMATICS AND COMPUTING↗

Demystifying asynchronous I/O Interference in HPC applications

With increasing complexity of HPC workflows, data management services need to perform expensive I/O operations asynchronously in the background, aiming to overlap the I/O with the application runtime. However, this may cause interference due to competition for resources: CPU, memory/network bandwidth. The advent of multi-core architectures has exacerbated this problem, as many I/O operations are issued concurrently, thereby competing not only with the application but also among themselves. Furthermore, the interference patterns can dynamically change as a response to variations in application behavior and I/O subsystems (e.g. multiple users sharing a parallel file system). Without a thorough understanding, I/O operations may perform suboptimally, potentially even worse than in the blocking case. To fill this gap, here we investigate the causes and consequences of interference due to asynchronous I/O on HPC systems. Specifically, we focus on multi-core CPUs and memory bandwidth, isolating the interference due to each resource. Then, we perform an in-depth study to explain the interplay and contention in a variety of resource sharing scenarios such as varying priority and number of background I/O threads and different I/O strategies: sendfile, read/write, mmap/write underlining trade-offs. The insights from this study are important both to enable guided optimizations of existing background I/O, as well as to open new opportunities to design advanced asynchronous I/O strategies.

97 MATHEMATICS AND COMPUTING↗

GPU-Accelerated Asynchronous Error Correction for Mixed Precision Iterative Refinement

In hardware-aware high performance computing, block- asynchronous iteration and mixed precision iterative refinement are two techniques that are applied to leverage the computing power of SIMD accelerators like GPUs. Although they use a very different approach for this purpose, they share the basic idea of compensating the convergence behaviour of an inferior numerical algorithm by a more efficient usage of the provided computing power. In this paper, we want to analyze the potential of combining both techniques. Therefore, we implement a mixed precision iterative refinement algorithm using a block-asynchronous iteration as an error correction solver, and compare its performance with a pure implementation of a block-asynchronous iteration and an iterative refinement method using double precision for the error correction solver. For matrices from theUniversity of FloridaMatrix collection,we report the convergence behaviour and provide the total solver runtime using different GPU architectures.

97 MATHEMATICS AND COMPUTING↗

A Fine-grained Asynchronous Bulk Synchronous parallelism model for PGAS applications

The Partitioned Global Address Space (PGAS) model is well suited for executing irregular applications on cluster-based systems, due to its efficient support for short, one-sided messages. Separately, the actor model has been gaining popularity as a productive asynchronous message-passing approach for distributed objects in enterprise and cloud computing platforms, typically implemented in languages such as Erlang, Scala or Rust. To the best of our knowledge, there has been no past work on using the actor model to deliver both productivity and scalability to irregular PGAS applications with large number of small messages. In this paper, we introduce a new programming system for PGAS applications, in which point-to-point remote operations can be expressed as fine-grained asynchronous actor messages. In our approach, the programmer does not need to worry about programming complexities related to message aggregation and termination detection. Our approach can be viewed as extending the classical Bulk Synchronous Parallelism model with fine-grained asynchronous communications within a phase or superstep. Here, we believe that our approach offers a desirable point in the productivity-performance space for PGAS applications, with more scalable performance and higher productivity relative to past approaches. Specifically, for seven irregular mini-applications from the Bale Kernels and three graph kernels executed using 2048 cores in the NERSC Cori system, our approach shows geometric mean performance improvements of ≥ 20X relative to standard PGAS versions (UPC and OpenSHMEM) while maintaining comparable productivity to those versions.

97 MATHEMATICS AND COMPUTING↗

Performance of asynchronous optimized Schwarz with one-sided communication

In asynchronous iterative methods on distributed-memory computers, processes update their local solutions using data from other processes without an implicit or explicit global synchronization that corresponds to advancing the global iteration counter. Here in this work, we test the asynchronous optimized Schwarz domain-decomposition iterative method using various one-sided (remote direct memory access) communication schemes with passive target completion. The results show that when one-sided communication is well-supported, the asynchronous version of optimized Schwarz can outperform the synchronous version even for perfectly balanced partitionings of the problem on a supercomputer with uniform nodes.

97 MATHEMATICS AND COMPUTING↗

Efficient Asynchronous I/O with Request Merging

With the advancement of exascale computing, the amount of scientific data is increasing day by day. Efficient data access is necessary for scientific discoveries. Unfortunately, the I/O performance is not improved, like the CPU and network speed. So, I/O operations take longer time than data generation or analysis. Asynchronous I/O has been proposed to extenuate the I/O bottleneck by overlapping I/O and computation time. However, multiple small write operations can diminish the benefits of asynchronous I/O, as the I/O time becomes significantly longer than the compute time, with little time to overlap with. To overcome these issues, we present an optimization technique to merge small contiguous write operations. We integrated our solution into the HDF5 asynchronous I/O VOL connector and demonstrated the effectiveness of merging HDF5 write operations automatically and transparently without requiring any code change from the application.

Chowdhury, Kamal Hossain↗

Accelerating Flash-X Simulations with Asynchronous I/O

Most high-fidelity physics simulation codes, such as Flash-X, need to save intermediate results (checkpoint files) to restart or gain insights into the evolution of the simulation. These simulation codes save such intermediate files synchronously, where computation is stalled while the data is written to storage. Depending on the problem size and computational requirements, this file write time can be a substantial portion of the total simulation time. In order to hide the I/O latency of checkpointing, asynchronous I/O methods have been introduced. These methods use background threads for performing I/O while the main threads continue with the simulation. The usage of background threads can compete for resources on the node as well as with communication. In this paper, we evaluate the overheads and the overall benefit of asynchronous I/O in HDF5 to simulations. Results from real-world high-fidelity simulations on the Summit supercomputer show that I/O operation is overlapped with application communication or computation or both, effectively hiding some or all of the I/O latency. Our evaluation shows that while using asynchronous I/O adds overhead to the application, the I/O time reduction is more significant, resulting in overall up to 1.5X performance speedup.

Jain, Rajeev↗

Divergence Reduction in Monte Carlo Neutron Transport with On-GPU Asynchronous Scheduling

While Monte Carlo Neutron Transport (MCNT) is near-embarrasingly parallel, the effectively unpredictable lifetime of neutrons can lead to divergence when MCNT is evaluated on GPUs. Divergence is the phenomenon of adjacent threads in a warp executing different control flow paths; on GPUS, it reduces performance because each work group may only execute one path at a time. The process of Thread Data Remapping (TDR) resolves these discrepancies by moving data across hardware such that data in the same warp will be processed through similar paths. A common issue among prior implementations of TDR is the synchronous nature of its remapping and processing cycles, which exhaustively sort data produced by prior processing passes and exhaustively evaluate the sorted data. In another work, we defined a method of remapping data through an asynchronous scheduler which allows for work to be stored in shared memory and deferred arbitrarily until that work is a viable option for low-divergence evaluation. This article surveys a wider set of cases, with the goal of characterizing performance trends across a more comprehensive set of parameters. These parameters include cross sections of scattering/capturing/fission, use of implicit capture, source neutron counts, simulation time spans, and tuned memory allocations. Across these cases, we have recorded minimum and average execution times, as well as a heuristically tuned near-optimal memory allocation size for both synchronous and asynchronous scheduling. Across the collected data, it is shown that the asynchronous method is faster and more memory efficient in the majority of cases, and that it requires less tuning to achieve competitive performance.

Computer Science↗

Asynchronous Stoichiometric Response in Lithium Iron Phosphate Batteries

The operando energy-dispersive x-ray diffraction (EDXRD) was carried out on a newly formed 8 Ah lithium iron phosphate (LiFePO4) battery with the goal of elucidating the origin of asynchronous phase transformation commonly seen with in situ x-ray diffraction studies. The high-energy photons at the NSLS X17B1 beamline allow for penetration into a fully assembled battery and therefore negate any need for a specially designed in situ cell which often uses modified current collectors to minimize x-ray attenuation. Spatially-and-temporally resolved phase-mapping was conducted with a semiquantitative reference intensity ratio (RIR) analysis to estimate the relative abundance of the delithiated phase. The data show an asynchronous response in the stoichiometry versus the electrochemical profile and suggest limited diffusion in the electrode toward the end of discharge. These results confirm that the asynchronous electrode response is not just limited to specially designed cells but occurs in fully assembled cells alike. We attribute this behavior to be a consequence of performing a local measurement over a wide-area heterogeneous reaction.

25 ENERGY STORAGE↗

Asynchronous data change notification between database server and accelerator controls system

Database data change notification (DCN) is a commonly used feature. Not all database management systems (DBMS) provide an explicit DCN mechanism. Even for those DBMS's which support DCN (such as Oracle and MS SQL server), some server side and/or client side programming may be required to make the DCN system work. This makes the setup of DCN between database server and interested clients tedious and time consuming. In accelerator control systems, there are many well established software client/server architectures (such as CDEV, EPICS, and ADO) that can be used to implement data reflection servers that transfer data asynchronously to any client using the standard SET/GET API. This paper describes a method for using such a data reflection server to set up asynchronous DCN (ADCN) between a DBMS and clients. This method works well for all DBMS systems which provide database trigger functionality. Asynchronous data change notification (ADCN) between database server and clients can be realized by combining the use of a database trigger mechanism, which is supported by major DBMS systems, with server processes that use client/server software architectures that are familiar in the accelerator controls community (such as EPICS, CDEV or ADO). This approach makes the ADCN system easy to set up and integrate into an accelerator controls system. Several ADCN systems have been set up and used in the RHIC-AGS controls system.

43 PARTICLE ACCELERATORS↗