Speculative data-parallel algorithms for language recognition have been widely experimented for various types of finitestate automata (FA), deterministic (DFA) and nondeterministic (NFA), often derived fromregular exp...
详细信息
ISBN:
(纸本)9798400714436
Speculative data-parallel algorithms for language recognition have been widely experimented for various types of finitestate automata (FA), deterministic (DFA) and nondeterministic (NFA), often derived fromregular expressions (RE). Such an algorithm cuts the input string into chunks, independently recognizes each chunk in parallel by means of identical FAs, and at last joins the chunk results and checks the overall consistency. In chunk recognition, it is necessary to speculatively start the FAs in any state, thus causing an overhead that reduces the speedup over a serial algorithm. the existing data-parallel DFA-based recognizers suffer from an excessive number of starting states, and the NFA-based ones suffer from the number of nondeterministic transitions.
Deep neural networks (DNNs) increasingly rely on parallel structures to enhance performance and efficiency. However, existing machine learning compilers (MLCs) face challenges in optimizing these structures due to lim...
详细信息
ISBN:
(纸本)9798400714436
Deep neural networks (DNNs) increasingly rely on parallel structures to enhance performance and efficiency. However, existing machine learning compilers (MLCs) face challenges in optimizing these structures due to limited parallel fusion scopes and insufficient consideration of intra-operator information. this paper introduces Magneto, a novel framework designed to accelerate parallel structures in DNNs through the co-optimization of parallel operators. By expanding the scope of parallel operator fusion and introducing a dedicated co-tuning algorithm, Magneto unlocks new opportunities for co-optimization. Experimental results demonstrate that Magneto outperforms NVIDIA TensorRT and AMD MIGraphX, achieving speedups of 3.02× and 4.19×, respectively.
Molecular dynamics simulation emerges as an important area that HPC+AI helps to investigate the physical properties, with machine-learning interatomic potentials (MLIPs) being used. General-purpose machine-learning (M...
详细信息
ISBN:
(纸本)9798400714436
Molecular dynamics simulation emerges as an important area that HPC+AI helps to investigate the physical properties, with machine-learning interatomic potentials (MLIPs) being used. General-purpose machine-learning (ML) tools have been leveraged in MLIPs, but they are not perfectly matched with each other, since many optimization opportunities in MLIPs have been missed by ML tools. this inefficiency arises from the fact that HPC+AI applications work with far more computational complexity compared with pure AI scenarios. this paper has developed an MLIP, named TensorMD, independently from any ML tool. TensorMD has been evaluated on two supercomputers and scaled to 51.8 billion atoms, i.e., ~ 3× compared with state-of-the-art.
作者：
PHILIPPSEN, MICSI
International Computer Science Institute Berkeley CA and Dept. of Informatics University of Karlsruhe
this paper investigates the problem of aligning array data and processes in a distributed-memory implementation. We present complete algorithms for compile-time analysis, the necessary program restructuring, and subse...
详细信息
ISBN:
(纸本)9780897917001
this paper investigates the problem of aligning array data and processes in a distributed-memory implementation. We present complete algorithms for compile-time analysis, the necessary program restructuring, and subsequent code-generation, and discuss their complexity. We finally evaluate the practical usefulness by quantitative experiments. the technique presented analyzes complete programs, including branches, loops, and nested parallelism. Alignment is determined with respect to offset, stride, and general ass's relations. Pplacement of both data and processes are computed in a unifying framework based on an extended preference graph and its analysis. Dynamic redistributions are derived. the experimental results are very encouraging. the optimization algorithms implemented in our Modula-2* compiler improved the execution times of the programs by an average over 40% on a MasPar MP-1 with 16384 processors.
It is widely acknowledged in high-performance computing circles that parallel input/output needs substantial improvement in order to make scalable computers truly usable. We present a data storage model that allows pr...
详细信息
It is widely acknowledged in high-performance computing circles that parallel input/output needs substantial improvement in order to make scalable computers truly usable. We present a data storage model that allows processors independent access to their own data and a corresponding compilation strategy that integrates data-parallel computation with data distribution for out-of-core problems. Our results compare several communication methods and I/O optimizations using two out-of-core problems, Jacobi iteration and LU factorization.
In this paper, we evaluate the performance and usability of the parallelprogramming model OpenMP Superscalar (OmpSs), apply it to 10 different benchmarks and compare its performance with corresponding POSIX threads i...
详细信息
ISBN:
(纸本)9781450311601
In this paper, we evaluate the performance and usability of the parallelprogramming model OpenMP Superscalar (OmpSs), apply it to 10 different benchmarks and compare its performance with corresponding POSIX threads implementations.
We present a tool for mesh-partitioning parallelization of numerical programs working iteratively on an unstructured mesh. this conventional method splits a mesh into sub-meshes, adding some overlap on the boundaries ...
详细信息
ISBN:
(纸本)9780897919067
We present a tool for mesh-partitioning parallelization of numerical programs working iteratively on an unstructured mesh. this conventional method splits a mesh into sub-meshes, adding some overlap on the boundaries of the sub-meshes. the program is then run in SPMD mode on a parallel architecture with distributed memory. It is necessary to add calls to communication routines at a few carefully selected locations in the code. the tool presented here uses the data-dependence information to mechanize the placement of these synchronizations. Additionally, we see that there is not a unique solution for placing these synchronizations, and performance depends on this choice.
暂无评论