期刊文献+
共找到317篇文章
< 1 2 16 >
每页显示 20 50 100
DNA-based方法及其在研究动物食物组成中的应用 被引量:2
1
作者 马秀慧 王志坚 《四川动物》 CSCD 北大核心 2012年第3期497-503,共7页
了解动物的食物组成是研究生态系统的多样性和生态系统功能的基本要求,在动物生态学的研究中至关重要。DNA-based方法包括样本采集、DNA提取、PCR扩增、序列比对分析等一系列步骤,实验的成败取决于标记基因的选择、引物的设计及每个步... 了解动物的食物组成是研究生态系统的多样性和生态系统功能的基本要求,在动物生态学的研究中至关重要。DNA-based方法包括样本采集、DNA提取、PCR扩增、序列比对分析等一系列步骤,实验的成败取决于标记基因的选择、引物的设计及每个步骤的优化。文章综述了DNA-based方法的各个基本步骤及其优化方法,并对该方法在今后用于国内相关研究的可能性进行展望,关注用于研究国内三峡大坝这个大型水域生态系统的可能性,期望推动国内该领域的研究。 展开更多
关键词 dna-based方法 食物 目标基因 生态系统 三峡大坝
在线阅读 下载PDF
Integrated DNA-based/biochemical screening for early diagnosis of multiple endocrine neoplasia type 2A (MEN2A) 被引量:3
2
作者 Qin Cui Wen Wang +7 位作者 Zhenzhen Fu Xin Shao Zhihong Zhang Mei Zhang Xianxia Ju Kunlin Wang Jiawei Chen Hongwen Zhou 《The Journal of Biomedical Research》 CAS 2013年第2期145-150,共6页
Multiple endocrine neoplasia type 2A (MEN2A), a subtype of MEN2, is characterized by medullary thyroid cancer, pheochromocytoma, and primary hyperparathyroidism. A Han Chinese pedigree with MEN2A was investigated fo... Multiple endocrine neoplasia type 2A (MEN2A), a subtype of MEN2, is characterized by medullary thyroid cancer, pheochromocytoma, and primary hyperparathyroidism. A Han Chinese pedigree with MEN2A was investigated following confirmation of the proband's diagnosis by pathological findings and DNA/biochemical screening. DNA samples from 4 other family members were collected and exon 5, 8, 10, 11, 13, 16 and 18 of the RET proto-oncogene were sequenced and then analyzed. A missense mutation of TGG (Trp) to TGC (Cys) at codon 634 (the classic MEN2A mutation) in exon 11 of the RET gene was detected in 3 family members, including the proband. Sequencing data were compared with the human gene mutation database. Elevated serum calcitonin level was detected initially; medullary thyroid carcinoma was revealed in the 3 cases and adrenal pheochromocytoma was also found in the proband. Elective operations were successfully performed on the adrenal and thyroid glands because of pheochromocytoma and medullary thyroid carcinoma. Our case study confirms that integrated DNA-based/biochemical screening is crucial for early diagnosis of MEN2A and is helpful in the screening of their relatives. In addition, DNA-based screening may occasionally uncover a previously unknown RET sequence. 展开更多
关键词 MEN-2A RET mutation medullary thyroid carcinoma Chinese RET proto-oncogene Calcitoninscreening dna-based screening
暂未订购
An ionic liquid supported CeO_2 nanoparticles-carbon nanotubes composite-enhanced electrochemical DNA-based sensor for the detection of Pb^(2+) 被引量:2
3
作者 Yan Lia Xiao-Rong Liut +3 位作者 Xiao-Hui Ning Can-Can Hnang Jian-Bin Zheng Jun-cai Zhang 《Journal of Pharmaceutical Analysis》 SCIE CAS 2011年第4期258-263,共6页
An electrochemical sensor incorporating a signal enhancement for the determination of lead (II) ions (Pb2+) was designed on the basis of the thrombin-binding aptamer (TBA) as a molecular recog- nition element a... An electrochemical sensor incorporating a signal enhancement for the determination of lead (II) ions (Pb2+) was designed on the basis of the thrombin-binding aptamer (TBA) as a molecular recog- nition element and ionic liquid supported cerium oxide (CeO2) nanoparticles-carbon nanotubes compo- site modification. The composite comprises nanoparticles CeO2, multi-waU carbon nanotubes (MWNTs) and hydrophobic room temperature ionic liquid (RTIL) 1-ethyl-3-methylimidazolium tetrafluoroborate (EMIMBF4). The electrochemical sensors were fabricated by immersing the CeOa-MWNTs-EMIMBF4 modified glassy carbon electrode (GCE) into the solution of TBA probe. In the presence of Pb2+, the TBA probe could form stable G-quartet structure by the specific binding interactions between Pb2+ and TBA. The TBA-bound Pb2+ can be electrochemically reduced, which provides a readout signal for quantitative detection of Pb2+. The reduction peak current is linearly related to the concentration of Pb2+ from 1.0 * 10-8 M to 1.0 * 105 M with a detection limit of 5 * 109 M. This work demonstrates that the CeOz-MWNTs-EMIMBF4 nanocomposite modified GCE provides a promising platform for immobi- lizing the TBA probe and enhancing the sensitivity of the DNA-based sensors. 展开更多
关键词 ELECTROCHEMICAL dna-based sensor Lead (II) ion Ionic liquid Ce02 nanoparticle Multi-wall Carbonnanotubes
在线阅读 下载PDF
Typhoon Case Comparison Analysis Between Heterogeneous Many-Core and Homogenous Multicore Supercomputing Platforms
4
作者 LIU Xin YU Xiaolin +5 位作者 ZHAO Haoran HAN Qiqi ZHANG Jie WANG Chengzhi MA Weiwei XU Da 《Journal of Ocean University of China》 SCIE CAS CSCD 2023年第2期324-334,共11页
In this paper,a typical experiment is carried out based on a high-resolution air-sea coupled model,namely,the coupled ocean-atmosphere-wave-sediment transport(COAWST)model,on both heterogeneous many-core(SW)and homoge... In this paper,a typical experiment is carried out based on a high-resolution air-sea coupled model,namely,the coupled ocean-atmosphere-wave-sediment transport(COAWST)model,on both heterogeneous many-core(SW)and homogenous multicore(Intel)supercomputing platforms.We construct a hindcast of Typhoon Lekima on both the SW and Intel platforms,compare the simulation results between these two platforms and compare the key elements of the atmospheric and ocean modules to reanalysis data.The comparative experiment in this typhoon case indicates that the domestic many-core computing platform and general cluster yield almost no differences in the simulated typhoon path and intensity,and the differences in surface pressure(PSFC)in the WRF model and sea surface temperature(SST)in the short-range forecast are very small,whereas a major difference can be identified at high latitudes after the first 10 days.Further heat budget analysis verifies that the differences in SST after 10 days are mainly caused by shortwave radiation variations,as influenced by subsequently generated typhoons in the system.These typhoons generated in the hindcast after the first 10 days attain obviously different trajectories between the two platforms. 展开更多
关键词 heterogeneous many-core supercomputing platform homogenous multicore supercomputing platform comparison analysis typhoon case
在线阅读 下载PDF
Research advancement of DNA-based intelligent hydrogels:Manufacture, characteristics, application of disease diagnosis and treatment
5
作者 Zhe-Han Yang Jie Yin +4 位作者 Lei Xin Yuanfang Li Yijie Huang Ruo Yuan Ying Zhuo 《Chinese Chemical Letters》 SCIE CAS CSCD 2024年第10期92-101,共10页
DNA-based hydrogels are exceptional materials for biological applications because of their numerous advantages such as biodegradability,biocompatibility,hydrophilicity,super absorbency,porosity,and swelling.Among thes... DNA-based hydrogels are exceptional materials for biological applications because of their numerous advantages such as biodegradability,biocompatibility,hydrophilicity,super absorbency,porosity,and swelling.Among these advantages,the ability of DNA-based hydrogels to respond to specific physical and chemical triggers and undergo reversible phase transitions has garnered significant attention in the fields of disease diagnosis(biosensors)and treatment(drug delivery).This article focuses on the recent advancements in the research of DNA-based hydrogels and discusses the different types of these hydrogels,the synthetic methods,their unique properties,and their applications in biosensors and drug delivery.The types of DNA hydrogels are categorized based on their building blocks,and the process of synthesis as well as the unique characteristics of DNA-based hydrogels are described.Then,DNA-based responsive hydrogels utilized as intelligent materials for the development of biosensors are reviewed.Furthermore,this article also presents the current status of DNA-based responsive hydrogels in drug delivery for cancer treatment,wound healing,and other therapeutic applications.Ultimately,this paper discusses the current challenges in expanding the practical application of DNA-based hydrogels. 展开更多
关键词 dna-based hydrogel Intelligent hydrogel MANUFACTURE CHARACTERISTICS Diseasediagnosis and treatment
原文传递
Design of Queue for DNA-based Computer
6
作者 李汪根 丁永生 《Journal of Donghua University(English Edition)》 EI CAS 2006年第2期47-50,共4页
This paper discusses the design of the queue for DNA-based computer on the point view of data structure. The nucleotide encodings for all components of the queue are given out formally. The linear double-stranded DNA ... This paper discusses the design of the queue for DNA-based computer on the point view of data structure. The nucleotide encodings for all components of the queue are given out formally. The linear double-stranded DNA molecules are used as the storage structure of the queue, and the basic bio-operations over the queue are described. Furthermore, the comparison between the queue of the electronic computer and that of DNA-based computer are elucidated. To prove the feasibility of our work, nucleotide encodings for an instance of queue are given out. All the biological technology mentioned in this paper can be practically implemented in the laboratory. Based on this work, other data structures could be developed in DNA-based computer. 展开更多
关键词 data structure QUEUE dna-based computer.
在线阅读 下载PDF
SW-DDFT: Parallel Optimization of the Dynamical Density Functional Theory Algorithm Based on Sunway Bluelight II Supercomputer
7
作者 Xiaoguang Lv Tao Liu +5 位作者 Han Qin Ying Guo Jingshan Pan Dawei Zhao Xiaoming Wu Meihong Yang 《Computers, Materials & Continua》 2025年第7期1417-1436,共20页
The Dynamical Density Functional Theory(DDFT)algorithm,derived by associating classical Density Functional Theory(DFT)with the fundamental Smoluchowski dynamical equation,describes the evolution of inhomo-geneous flui... The Dynamical Density Functional Theory(DDFT)algorithm,derived by associating classical Density Functional Theory(DFT)with the fundamental Smoluchowski dynamical equation,describes the evolution of inhomo-geneous fluid density distributions over time.It plays a significant role in studying the evolution of density distributions over time in inhomogeneous systems.The Sunway Bluelight II supercomputer,as a new generation of China’s developed supercomputer,possesses powerful computational capabilities.Porting and optimizing industrial software on this platform holds significant importance.For the optimization of the DDFT algorithm,based on the Sunway Bluelight II supercomputer and the unique hardware architecture of the SW39000 processor,this work proposes three acceleration strategies to enhance computational efficiency and performance,including direct parallel optimization,local-memory constrained optimization for CPEs,and multi-core groups collaboration and communication optimization.This method combines the characteristics of the program’s algorithm with the unique hardware architecture of the Sunway Bluelight II supercomputer,optimizing the storage and transmission structures to achieve a closer integration of software and hardware.For the first time,this paper presents Sunway-Dynamical Density Functional Theory(SW-DDFT).Experimental results show that SW-DDFT achieves a speedup of 6.67 times within a single-core group compared to the original DDFT implementation,with six core groups(a total of 384 CPEs),the maximum speedup can reach 28.64 times,and parallel efficiency can reach 71%,demonstrating excellent acceleration performance. 展开更多
关键词 Sunway supercomputer high-performance computing dynamical density functional theory parallel optimization
在线阅读 下载PDF
More Super Supercomputers 被引量:2
8
作者 Jane Palmer 《Engineering》 SCIE EI 2019年第3期357-358,共2页
In June 2018, the United States claimed the No. 1 position in supercomputing according to TOP500, which ranks the top 500 most powerful computer systems in the world [1]. The US Department of Energy’s Summit machine ... In June 2018, the United States claimed the No. 1 position in supercomputing according to TOP500, which ranks the top 500 most powerful computer systems in the world [1]. The US Department of Energy’s Summit machine (Fig. 1)[1] claimed this distinction, which previously had been held by China’s Sunway TaihuLight supercomputer. 展开更多
关键词 supercomputERS the UNITED STATES China
在线阅读 下载PDF
World’s Fastest Supercomputer Picks COVID-19 Drug
9
作者 Zameer Shervani Intazam Khan +1 位作者 Tasrina Khan Umair Yaqub Qazi 《Advances in Infectious Diseases》 2020年第3期211-225,共15页
We have demonstrated the application of the world’s fastest supercomputer Fugaku located in Japan to select the COVID-19 drugs and stopping the pandemic spread. Using computer simulation out of 2128 potential drug ca... We have demonstrated the application of the world’s fastest supercomputer Fugaku located in Japan to select the COVID-19 drugs and stopping the pandemic spread. Using computer simulation out of 2128 potential drug candidates, the world’s fastest supercomputer picked 30 most effective and potential drugs. Twelve of them are under clinical trials outside Japan;some are being tested in Japan. The computer reduced the computation time from one year to 10 days when compared to second superfast computer of the world. Fugaku supercomputer was employed to know the behavior of airborne aerosol COVID-19 virus. 3Cs were suggested: avoid closed and crowded spaces and contacts to stop the pandemic spread. The progress in vaccine development and proper use and type of mask has also been described in this article. The article will benefit greatly to stop spreading and treating the pandemic COVID-19. 展开更多
关键词 COVID-19 Drug World’s Fastest supercomputer Fugaku supercomputer COVID-19 Treatment COVID-19 Prevention
在线阅读 下载PDF
The Chinese Supercomputer
10
《Rare Metals》 SCIE EI CAS CSCD 1993年第1期71-72,共2页
The first in China 10~9 sparallel supercomputer, named as Yinhe-Ⅱ, had been manufac-tured by Science-technological University of National Defence. The main feature of thesupercomputer are: 4-processor system, the pri... The first in China 10~9 sparallel supercomputer, named as Yinhe-Ⅱ, had been manufac-tured by Science-technological University of National Defence. The main feature of thesupercomputer are: 4-processor system, the principle frequence 50 MHz, the word length 64 byte,the main memory 256 Mb, two individual input / output subsystems, > 10~9 operations per sec- 展开更多
关键词 LENGTH The Chinese supercomputer
在线阅读 下载PDF
The Development and Innovation of the First Supercomputer YH-1 in China:From the Perspective of Technology History
11
作者 SI Hongwei FENG Lisheng 《Chinese Annals of History of Science and Technology》 2021年第2期161-186,共26页
China’s first supercomputer capable of 100 million calculations per second was the YH-1,which was independently developed by the Institute of Computer Science at the National University of Defense Technology(NUDT)bet... China’s first supercomputer capable of 100 million calculations per second was the YH-1,which was independently developed by the Institute of Computer Science at the National University of Defense Technology(NUDT)between 1978 and 1983.YH-1 played an important role in China’s national defense construction and national economic development.It made China one of the few countries in the world to successfully develop a supercomputer.Based on original archive documents,interviews with relevant personnel,and an analysis of the technological parameters of the supercomputers YH-1 in China and Cray-1 in the United States,this paper reviews in detail the historic process of the development of YH-1,analyzing its innovation and summarizing the experience and lessons learned from it.This analysis is significant for current military-civilian integration,and the commercialization of university research findings in China. 展开更多
关键词 history of technology supercomputER YH-1 development and innovation
在线阅读 下载PDF
太空超算:概念、挑战及应用
12
作者 龚春叶 董皓 +4 位作者 包为民 李超 朱浩 王之 孟祥飞 《宇航学报》 北大核心 2025年第7期1263-1273,共11页
太空超算是指在地球外太空建立的超级计算系统,具备科学计算、智能计算和大数据处理能力。随着“后摩尔时代”的到来,以及能源技术与航天技术的快速发展,面对人类太空移民的迫切需求,太空超算有望成为引领未来科技发展的战略制高点。太... 太空超算是指在地球外太空建立的超级计算系统,具备科学计算、智能计算和大数据处理能力。随着“后摩尔时代”的到来,以及能源技术与航天技术的快速发展,面对人类太空移民的迫切需求,太空超算有望成为引领未来科技发展的战略制高点。太空超算作为新概念产品,面临着体系架构技术、兆瓦级太空发电及储能技术、散热技术、通信技术、可靠性技术、运维技术、可重复使用运载技术、低成本技术和可持续运营等挑战。太空超算在支撑天数天算和太空基地运营,牵引天电天用、可重复使用运载及太空信息技术发展等方面发挥着重要应用价值。预计投资数十亿元量级,可在未来5年左右建成P级太空超算验证系统,30年左右形成产业化运营;人类将从卫星、空间站时代进入太空信息时代,对太空的开发和利用将登上新台阶。 展开更多
关键词 超级计算 太空超算 太空电力 太空基地 太空信息
在线阅读 下载PDF
超算和智算算力池化系统研究和实践 被引量:1
13
作者 杨振东 刘沛强 +1 位作者 冯铭能 陈善杰 《通信与信息技术》 2025年第3期87-90,共4页
提出一种超算和智算算力池化系统,旨在解决当前算力应用市场中存在的注册、充值流程烦琐、使用门槛高等问题。通过深度集成多个算力中心,建立统一的算力门户,用户可线上注册、充值并使用多个算力中心的服务,实现即开即用。同时,通过算... 提出一种超算和智算算力池化系统,旨在解决当前算力应用市场中存在的注册、充值流程烦琐、使用门槛高等问题。通过深度集成多个算力中心,建立统一的算力门户,用户可线上注册、充值并使用多个算力中心的服务,实现即开即用。同时,通过算力池化系统,将算力中心的2B业务转化为2C模式,降低充值和消费门槛,推动超算和智算算力应用的普及。详细阐述了系统架构和技术方案,包括算力控制台、用户管理、算力池管理模块等,为算力服务的高效对接提供了创新解决方案。 展开更多
关键词 超算 智算 算力资源整合 算力池
在线阅读 下载PDF
浅析天河图计算
14
作者 甘新标 熊锋 《计算》 2025年第4期67-74,共8页
图计算是一种专用于处理图结构数据的计算模型,广泛应用于社交网络、交通优化及军事等领域。本文概述了图计算的基本概念与核心任务,并回顾其发展历程。重点介绍了国防科技大学研发的“天河图计算系统”(TianheGraph),其通过软硬协同图... 图计算是一种专用于处理图结构数据的计算模型,广泛应用于社交网络、交通优化及军事等领域。本文概述了图计算的基本概念与核心任务,并回顾其发展历程。重点介绍了国防科技大学研发的“天河图计算系统”(TianheGraph),其通过软硬协同图分布(GraphCube)、拓扑感知通信MST和超级图存储(SuperCSR)等关键技术,显著提升了大规模图数据处理的效率与性能。天河图计算系统在Graph500排名中多次夺冠,为各行业赋能,展现了国产超算在图计算领域的领先地位。未来,图计算将与人工智能融合,推动更广泛的应用创新。 展开更多
关键词 图计算 分布式计算 图存储 图通信 超级计算机 Graph500
在线阅读 下载PDF
AI4S的若干思考
15
作者 孙凝晖 《计算》 2025年第5期16-19,共4页
人工智能(artificial intelligence,AI)正在催生新一轮科技革命,其中人工智能赋能科学研究(AIforScience,AI4S)为人类探索未知开启了一种新的科研范式。本文讨论了AI如何提高科研活动的劳动生产率,助力突破人类认知极限的典型技术路线,... 人工智能(artificial intelligence,AI)正在催生新一轮科技革命,其中人工智能赋能科学研究(AIforScience,AI4S)为人类探索未知开启了一种新的科研范式。本文讨论了AI如何提高科研活动的劳动生产率,助力突破人类认知极限的典型技术路线,探讨了AI4S中间件支撑平台的共性功能,以及AI4S普及教育的相关思考。 展开更多
关键词 AI4S 科研范式 智能超算 科学基础大模型 中间件 模型工厂
在线阅读 下载PDF
基于超级计算机的高性能计算应用发展现状及趋势研究 被引量:2
16
作者 刘扬 许建飞 +9 位作者 许黄超 吴璨 胡泰源 原惠峰 高凌云 梁文昊 董盛 马英晋 李瑞琳 赵永华 《数据与计算发展前沿(中英文)》 2025年第2期68-85,共18页
【目的】随着信息技术的快速发展和全球数据量的激增,超级计算机(超算)已经成为科学研究和创新发展的重要驱动力。本文旨在探讨超算在多个领域中的应用现状与发展趋势。【方法】通过广泛调研全球范围内的超算和领域应用情况,系统性地对... 【目的】随着信息技术的快速发展和全球数据量的激增,超级计算机(超算)已经成为科学研究和创新发展的重要驱动力。本文旨在探讨超算在多个领域中的应用现状与发展趋势。【方法】通过广泛调研全球范围内的超算和领域应用情况,系统性地对相关高性能计算应用进行分类和总结,重点关注化学与材料、物理学等多个领域,探讨相关计算需求与超算的适配和部署情况。此外,本文还积极讨论了网格计算与超算互联。【结果】超算在多个领域应用已经展现出了显著的效果。随着应用领域的需要和高性能计算技术的不断发展,对超级计算机的软硬件发展也提出更高要求。【局限】虽然超算正处在蓬勃发展的阶段,可应用范围广泛,但本文仅选取了代表性应用领域进行分析总结。【结论】超算在加速科学发现和技术创新方面的效率显著提升,为未来的研究和应用提供了强有力的支持。同时,提升超算的性能和适应性将是未来科研进展的重要保障。 展开更多
关键词 超级计算机 大规模并行应用 高性能计算
在线阅读 下载PDF
MTTorch:面向MT-3000芯片和Transformer模型的PyTorch算子库实现与优化 被引量:1
17
作者 王昊天 孙羽菲 +4 位作者 隋轶丞 王嘉豪 石昌青 方建滨 张玉志 《软件学报》 北大核心 2025年第8期3896-3916,共21页
随着Transformer类大模型的飞速发展,算力逐渐成为制约领域发展的瓶颈,如何根据加速器硬件的结构特性加速和优化大语言模型的训练性能已成为研究热点.面向天河新一代超算系统的加速芯片MT-3000,提出并实现了适用于CPU+DSP异构架构的PyTo... 随着Transformer类大模型的飞速发展,算力逐渐成为制约领域发展的瓶颈,如何根据加速器硬件的结构特性加速和优化大语言模型的训练性能已成为研究热点.面向天河新一代超算系统的加速芯片MT-3000,提出并实现了适用于CPU+DSP异构架构的PyTorch扩展库——MTTorch,其核心是一个多核并行的算子库,对Transformer类模型训练过程中的核心算子进行向量化实现和优化.同时,针对MT-3000架构特性,提出了面向多核DSP的高性能规约算法及乒乓算法,显著提升了算子的运算性能.MTTorch还具有很好的通用性,对于不同版本的PyTorch都可以动态链接库的形式进行加载,不改变PyTorch的原生实现.大量实验证明,实现的核心算子在MT-3000芯片上有着很好的性能,在单DSP簇上可以达到8倍的加速效果.利用MTTorch在多节点执行训练任务时有着接近线性的加速比,极大地提升了Transformer类模型在MT-3000芯片上的训练效率. 展开更多
关键词 PyTorch 高性能计算 Transformer模型 天河超级计算机 CPU+DSP异构计算 软件生态
在线阅读 下载PDF
算力基础设施建设能破解“生产率悖论”吗——基于合作创新和劳动力技能结构的视角 被引量:9
18
作者 杨壮 吴福象 《经济学家》 北大核心 2025年第1期56-65,共10页
算力基础设施是新型信息基础设施的重要组成部分,能够为企业发展提供高水平算力支撑,但在“生产率悖论”得到广泛验证的背景下,算力基础设施建设能否提升企业全要素生产率(TFP)是个值得研究的问题。本文以国家超级计算中心投入运营为外... 算力基础设施是新型信息基础设施的重要组成部分,能够为企业发展提供高水平算力支撑,但在“生产率悖论”得到广泛验证的背景下,算力基础设施建设能否提升企业全要素生产率(TFP)是个值得研究的问题。本文以国家超级计算中心投入运营为外生政策冲击,利用沪深A股上市公司数据和双重差分模型,探究算力基础设施建设对企业TFP的影响效应。研究发现,算力基础设施建设能够显著促进企业TFP提升,并且提升效应与企业智能化水平密切相关。作用机制检验表明,算力基础设施建设主要通过促进企业合作创新和推动企业劳动力技能结构调整两个渠道影响企业TFP。基于地区、行业、企业三个层面的异质性分析发现,算力基础设施建设对企业TFP的影响在发展禀赋较好地区、高技术行业和国有企业中更为明显。 展开更多
关键词 基础设施 国家超级计算中心 全要素生产率 “生产率悖论”
原文传递
swJulia:面向新一代神威超级计算机的Julia语言编译系统
19
作者 沈莉 周文浩 +5 位作者 王飞 李斌 谭坚 商红慧 安虹 漆锋滨 《软件学报》 北大核心 2025年第12期5402-5422,共21页
随着异构融合体系结构在高性能计算领域的普及,挖掘其潜能并探索新的应用构建策略变得至关重要.传统的静态编译方法已无法满足复杂计算需求,动态编程语言因其灵活性和高效性而备受瞩目.Julia是一种现代的高性能动态编程语言,其基于即时... 随着异构融合体系结构在高性能计算领域的普及,挖掘其潜能并探索新的应用构建策略变得至关重要.传统的静态编译方法已无法满足复杂计算需求,动态编程语言因其灵活性和高效性而备受瞩目.Julia是一种现代的高性能动态编程语言,其基于即时编译机制,在科学计算等领域表现出色.结合申威异构众核架构特点,构建ORCJIT编译引擎并提出了动态模式下的片上存储管理方法,并以此为基础实现针对新一代神威超级计算机的Julia动态语言编译器swJulia.其不仅继承了Julia编译器的灵活性,同时还有效支持了SACA众核编程模型及运行时封装.利用swJulia编译系统,成功在新一代神威超级计算机上部署了NNQS-Transformer量子化学模拟器,并在多个维度验证了swJulia的好用性和高效性.实验结果显示,swJulia在单线程基准测试和众核加速上性能卓越,并能够有效支撑NNQS-Transformer量子化学模拟器的超大规模可扩展并行模拟. 展开更多
关键词 Julia编译器 神威超级计算机 动态链接 即时编译 量子化学模拟
在线阅读 下载PDF
一种集成于超算作业调度系统应用的并行参数优化方法
20
作者 张文帅 李会民 +1 位作者 李京 潘必才 《计算机工程》 北大核心 2025年第7期59-67,共9页
随着高性能计算体系结构的发展,软件与硬件都具有多层的并行结构。当不同纵向层级与横向分组的计算任务被划分到不同节点的不同处理器时,存在非常多的分配方式。这些分配方式一般在运行时由用户输入的多个并行参数来确定,并对计算效率... 随着高性能计算体系结构的发展,软件与硬件都具有多层的并行结构。当不同纵向层级与横向分组的计算任务被划分到不同节点的不同处理器时,存在非常多的分配方式。这些分配方式一般在运行时由用户输入的多个并行参数来确定,并对计算效率影响很大。随着计算规模与复杂度的提升,多个并行参数的可配置空间越来越大,用户越来越难以确定最佳的并行参数值。这类运行时优化问题在科学计算应用中较为普遍,但相关的研究与解决方法比较少见。以VASP(Vienna Ab initio Simulation Package)应用为例,首先分析了该应用的多层并行结构,展示了不同并行参数配置引发的巨大运行速度差异。然后提出了一个基于约化并行效率指标的全自动运行优化方法,其不仅可以帮助用户简单快捷地确定最佳应用并行参数,而且可以帮助用户确定最佳的计算资源使用量,使应用可以高效率地扩展到大规模的并行计算中。最后将该优化方法与计算集群作业调度系统相融合应用于用户提交的真实VASP计算作业。统计结果表明,该方法显著提升了作业运行速度与超算资源的使用效率,具有很好的工程应用前景。 展开更多
关键词 并行计算 作业调度 运行时优化 超级计算 VASP应用
在线阅读 下载PDF
上一页 1 2 16 下一页 到第
使用帮助 返回顶部