检索结果-内蒙古大学图书馆

39th International Conference on High Performance computing, ISC High Performance 2024

作者： Nichols, Daniel Marathe, Aniruddha Menon, Harshitha Gamblin, Todd Bhatele, Abhinav Department of Computer Science University of Maryland College ParkMD United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory LivermoreCA United States Livermore Computing Lawrence Livermore National Laboratory LivermoreCA United States

ISBN: (纸本)9783982633602

Parallel programs in high performance computing (HPC) continue to grow in complexity and scale in the exascale era. The diversity in hardware and parallel programming models make developing, optimizing, and maintaining parallel software even more burdensome for developers. One way to alleviate some of these burdens is with automated development and analysis tools. Such tools can perform complex and/or remedial tasks for developers that increase their productivity and decrease the chance for error. Until recently, such tools for code development and performance analysis have been limited in the complexity of tasks they can perform, especially for parallel programs. However, with recent advancements in language modeling, and the availability of large amounts of open-source code related data, these tools have started to utilize predictive language models to automate more complex tasks. In this paper, we show how large language models (LLMs) can be applied to tasks specific to high performance and scientific codes. We introduce a new dataset of HPC and scientific codes and use it to fine-tune several pre-trained models. We compare several pre-trained LLMs on HPC-related tasks and introduce a new model, HPC-Coder, fine-tuned on parallel codes. In our experiments, we show that this model can auto-complete HPC functions where generic models cannot, decorate for loops with OpenMP pragmas, and model performance changes in scientific application repositories as well as programming competition solutions. © 2024 Research Paper proceedings of the ISC High Performance 2024. All rights reserved.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Thrilled by Your Progress! Large Language models (GPT-4) No Longer Struggle to Pass Assessments in Higher Education programming Courses 23

Thrilled by Your Progress! Large Language Models (GPT-4) No ...

引用

19th Annual ACM Conference on International computing Education Research V.1 (ICER)

作者： Savelka, Jaromir Agarwal, Arav An, Marshall Bogart, Chris Sakr, Majd Carnegie Mellon Univ Pittsburgh PA 15213 USA

ISBN: (纸本)9781450399760

This paper studies recent developments in large language models' (LLM) abilities to pass assessments in introductory and intermediate Python programming courses at the postsecondary level. The emergence of ChatGPT resulted in heated debates of its potential uses (e.g., exercise generation, code explanation) as well as misuses in programming classes (e.g., cheating). Recent studies show that while the technology performs surprisingly well on diverse sets of assessment instruments employed in typical programming classes the performance is usually not sufficient to pass the courses. The release of GPT-4 largely emphasized notable improvements in the capabilities related to handling assessments originally designed for human test-takers. This study is the necessary analysis in the context of this ongoing transition towards mature generative AI systems. Specifically, we report the performance of GPT-4, comparing it to the previous generations of GPT models, on three Python courses with assessments ranging from simple multiple-choice questions (no code involved) to complex programming projects with code bases distributed into multiple files (599 exercises overall). Additionally, we analyze the assessments that were not handled well by GPT-4 to understand the current limitations of the model, as well as its capabilities to leverage feedback provided by an auto-grader. We found that the GPT models evolved from completely failing the typical programming class' assessments (the original GPT-3) to confidently passing the courses with no human involvement (GPT-4). While we identified certain limitations in GPT-4's handling of MCQs and coding exercises, the rate of improvement across the recent generations of GPT models strongly suggests their potential to handle almost any type of assessment widely used in higher education programming courses. These findings could be leveraged by educators and institutions to adapt the design of programming assessments as well as to fuel the nec

关键词： AI code generation introductory and intermediate programming Multiple-choice question answering MCQ coding exercises generative pre-trained transformers GPT Python course programming knowledge assessment ChatGPT Codex GitHub Copilot Alpha-Code

来源：评论

学校读者我要写书评

暂无评论

programming big data analysis: principles and solutions

引用

JOURNAL OF BIG DATA 2022年第1期9卷 1-50页

作者： Belcastro, Loris Cantini, Riccardo Marozzo, Fabrizio Orsino, Alessio Talia, Domenico Trunfio, Paolo Univ Calabria Arcavacata Di Rende Italy Dtok Lab Arcavacata Di Rende Italy

In the age of the Internet of Things and social media platforms, huge amounts of digital data are generated by and collected from many sources, including sensors, mobile devices, wearable trackers and security cameras. This data, commonly referred to as Big Data, is challenging current storage, processing, and analysis capabilities. New models, languages, systems and algorithms continue to be developed to effectively collect, store, analyze and learn from Big Data. Most of the recent surveys provide a global analysis of the tools that are used in the main phases of Big Data management (generation, acquisition, storage, querying and visualization of data). Differently, this work analyzes and reviews parallel and distributed paradigms, languages and systems used today to analyze and learn from Big Data on scalable computers. In particular, we provide an in-depth analysis of the properties of the main parallel programming paradigms (MapReduce, workflow, BSP, message passing, and SQL-like) and, through programming examples, we describe the most used systems for Big Data analysis (e.g., Hadoop, Spark, and Storm). Furthermore, we discuss and compare the different systems by highlighting the main features of each of them, their diffusion (community of developers and users) and the main advantages and disadvantages of using them to implement Big Data analysis applications. The final goal of this work is to help designers and developers in identifying and selecting the best/appropriate programming solution based on their skills, hardware availability, application domains and purposes, and also considering the support provided by the developer community.

关键词： Parallel programming models programming systems Big Data analysis MapReduce Workflow Message Passing Bulk Synchronous Parallel SQL-like

来源：评论

学校读者我要写书评

暂无评论

SPLASH Companion 2023 - Companion proceedings of the 2023 ACM SIGPLAN International Conference on Systems, programming, languages, and Applications: Software for Humanity

SPLASH Companion 2023 - Companion Proceedings of the 2023 AC...

引用

38th ACM SIGPLAN International Conference on Systems, programming, languages and Applications: Software for Humanity, SPLASH 2023

ISBN: (纸本)9798400703843

The proceedings contain 26 papers. The topics discussed include: remote just-in-time compilation for dynamic languages;scaling up program synthesis to efficient algorithms;large language models for automated program repair;transforming ideas into code: visual sketching for ml development;semantic versioning for python programs;reusing single-language analyses for static analysis of multi-language programs;a pragmatic approach to syntax repair;partial gradual dependent type theory;synthesizing recursive programs through dataflow constraints;modular educational languages;historiographer: strongly-consistent distributed reactive programming with minimal locking;clearing the trail: motivations for maintenance work in open source;towards the formal verification of Wigderson’s algorithm;and involving users in design of a widely used language: a case of ECMAScript (JavaScript) standardization.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IRIS: Exploring Performance Scaling of the Intelligent Runtime System and its Dynamic Scheduling Policies

IRIS: Exploring Performance Scaling of the Intelligent Runti...

引用

1st International Conference on Smart Energy Systems and Artificial Intelligence (SESAI)

作者： Johnston, Beau Miniskar, Narasinga Rao Young, Aaron Monil, Mohammad Alaul Haque Lee, Seyong Vetter, Jeffrey S. Oak Ridge Natl Lab Oak Ridge TN 37830 USA

ISBN: (纸本)9798350364613;9798350364606

High-Performance computing is becoming increasingly heterogeneous, relying on a diverse mix of hardware to achieve good performance. Paradoxically, current drivers and frameworks for these devices typically require separate languages and implementations for each vendor. Furthermore, there are few tools and little support to schedule codes between these devices in a truly heterogeneous manner partly because of this fragmentation between vendors and the languages each supports. To overcome both limitations, the Intelligent Runtime System (IRIS) was developed. It allows a common task abstraction to automatically be shared among contemporary vendors and is run from a single host-side API. At runtime, IRIS queries the host system and registers which frameworks and drivers are available, these determine which kernels can be used by the scheduler-CPUs via OpenMP, Nvidia GPUs (CUDA), AMD CPUs (HIP), and Intel and Xilinx FPGAs with OpenCL. IRIS enables tasks to he scheduled to any heterogeneous device and resolves to the appropriate kernel binary at runtime it only uses the devices supported by the system on which it is run. IRIS supports single-task and graph-based expressions of dependencies of tasks. Additionally, IRIS features a range of dynamic scheduling policies, allowing complex chains of tasks and interactions to be executed, relieving the programmer/user from considering the system to assign tasks to devices optimally. This paper presents the peak performance attainable by IRIS over a range of systems each with different numbers and types of accelerator devices, it highlights the flexibility of IRIS since these devices are truly heterogeneous, relying on different back ends (drivers, frameworks, and languages) which historically required unique implementations to utilize them. We then use this peak performance as a baseline to compare increasingly complex chains of tasks (with increasingly complex task dependencies) and evaluate how IRIS copes. Finally, we consider

关键词： Execution Model Heterogeneous computing Heterogeneous Systems High-Performance computing Performance Portability programming models Runtime System Scheduling Policy Task Schedule

来源：评论

学校读者我要写书评

暂无评论

New Problems in distributed Inference for DNN models on Robotic IoT

New Problems in Distributed Inference for DNN Models on Robo...

引用

2024 Workshop on Advanced Tools, programming languages, and PLatforms for Implementing and Evaluating algorithms for distributed Systems, ApPLIED 2024, co-located with PODC 2024

作者： Sun, Zekai Guan, Xiuxian Wang, Junming Liu, Fangming Cui, Heming The University of Hong Kong Hong Kong Peng Cheng Laboratory Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9798400706707

The rapid advancements in machine learning (ML) techniques have led to significant achievements in various robotic tasks. Deploying these ML approaches on real-world robots requires fast and energy-efficient inference of their deep neural network (DNN) models. To our knowledge, distributed inference, which involves inference across multiple powerful GPU devices, has emerged as a promising optimization to improve inference performance in modern data centers. However, when deployed on real-world robots, existing parallel methods can not simultaneously meet the robots' latency and energy requirements and raise significant *** paper reveals and evaluates the problems hindering the application of these parallel methods in robotic IoT, including the failure of data parallelism, the unacceptable communication overhead of tensor parallelism, and the significant transmission bottlenecks in pipeline parallelism. By raising awareness of these new problems, we aim to stimulate research toward finding a new parallel method to achieve fast and energy-efficient distributed inference in robotic IoT. Copyright © 2024 held by the owner/author(s). Publication rights licensed to ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Safe Shared State in Dataflow Systems 18

Safe Shared State in Dataflow Systems

引用

18th International Conference on distributed and Event-Based Systems (DEBS)

作者： De Martini, Luca Margara, Alessandro Politecn Milan Milan Italy

ISBN: (纸本)9798400704437

The complexity of implementing scalable data processing tasks in parallel and distributed computing environments has pushed the adoption of restricted programming models that simplify expressing such tasks. In this context, the dataflow model has emerged as the standard for building distributed data processing systems. The key insight of this model is to express computations as sequences of operations that do not share any state and can thus be deployed and executed independently on the same or on different machines. however entirely excluding state sharing may be detrimental for performance, as it prevents operations executed on the same machine from accessing common resources. The effects of these limitations become more and more evident as the computational and memory resources of individual hosts increase, and even large-scale data analysis tasks can be performed with few well-equipped machines. Moving from this observation, in this paper we present an extension to the classic dataflow model that enables disciplined state sharing across operators to improve performance. We implement the model in Renoir, a dataflow system written in Rust, and we exploit the characteristics of this programming language to ensure sate access to shared state by design. We present several use cases where our model may be beneficial, and we use them to evaluate the proposed model in terms of performance and code complexity. We show that our model empowers developers to better exploit the resources of machines, resulting in substantial performance improvements in several use cases. Programs written with our extended model may even be more concise and simpler than in the classic dataflow model.

关键词： dataflow shared state distributed programming Rust memory safety data processing

来源：评论

学校读者我要写书评

暂无评论

Extensible 3D Simulation of Aggregated Systems with FCPP 24th

Extensible 3D Simulation of Aggregated Systems with FCPP

引用

24th IFIP WG 6.1 International Conference on Coordination models and languages (COORDINATION) Held as Part of the 17th International Federated Conference on distributed computing Techniques (DisCoTec)

作者： Audrito, Giorgio Rapetta, Luigi Torta, Gianluca Univ Torino Turin Italy

ISBN: (纸本)9783031081439;9783031081453

programming massively distributed systems in unreliable environments poses several non-trivial challenges. Such systems need to be able to adapt and self-organise, and special algorithms need to be developed for this purpose. In particular, simulators provide an irreplaceable tool for the development process. Among other tools for programming self-organizing systems, the FieldCalc++ (FCPP, implementing the field calculus in C++) library stands out for its efficiency, portability and extensibility, and its support for aggregate programs. On the other hand, the simulator's output was limited up to now to numeric statistical information, reducing the user's ability to understand and interact with the system under simulation. In this paper, we present a novel graphical user interface for FCPP, allowing for a real-time, interactive and three-dimensional visualization of the simulated system. Through this interface, the user can control the simulation flow, visualize summary information of the network at a single glance, and inspect detailed information via auxiliary windows. The interface is designed to require minimal effort from the end user for its setup, and can be further extended for increased interaction.

关键词： distributed computing Aggregate computing Toolchains

来源：评论

学校读者我要写书评

暂无评论

Locality in Online, Dynamic, Sequential, and distributed Graph Algorithms 50

Locality in Online, Dynamic, Sequential, and Distributed Gra...

引用

50th International Colloquium on Automata, languages, and programming, ICALP 2023

作者： Akbari, Amirreza Eslami, Navid Lievonen, Henrik Melnyk, Darya Särkijärvi, Joona Suomela, Jukka Aalto University Espoo Finland Sharif University of Technology Tehran Iran TU Berlin Germany

ISBN: (纸本)9783959772785

In this work, we give a unifying view of locality in four settings: distributed algorithms, sequential greedy algorithms, dynamic algorithms, and online algorithms. We introduce a new model of computing, called the online-LOCAL model: the adversary presents the nodes of the input graph one by one, in the same way as in classical online algorithms, but for each node we get to see its radius-T neighborhood before choosing the output. Instead of looking ahead in time, we have the power of looking around in space. We compare the online-LOCAL model with three other models: the LOCAL model of distributed computing, where each node produces its output based on its radius-T neighborhood, the SLOCAL model, which is the sequential counterpart of LOCAL, and the dynamic-LOCAL model, where changes in the dynamic input graph only influence the radius-T neighborhood of the point of change. The SLOCAL and dynamic-LOCAL models are sandwiched between the LOCAL and online-LOCAL models. In general, all four models are distinct, but we study in particular locally checkable labeling problems (LCLs), which is a family of graph problems extensively studied in the context of distributed graph algorithms. We prove that for LCL problems in paths, cycles, and rooted trees, all four models are roughly equivalent: the locality of any LCL problem falls in the same broad class – O(log∗ n), Θ(log n), or nΘ(1) – in all four models. In particular, this result enables one to generalize prior lower-bound results from the LOCAL model to all four models, and it also allows one to simulate e.g. dynamic-LOCAL algorithms efficiently in the LOCAL model. We also show that this equivalence does not hold in two-dimensional grids or general bipartite graphs. We provide an online-LOCAL algorithm with locality O(log n) for the 3-coloring problem in bipartite graphs – this is a problem with locality Ω(n1/2) in the LOCAL model and Ω(n1/10) in the SLOCAL model. © Amirreza Akbari, Navid Eslami, Henrik Lievonen, Darya

关键词： Computational complexity

来源：评论

学校读者我要写书评

暂无评论

Evaluation and Optimization of Sentiment Analysis Accuracy Based on Different models

Evaluation and Optimization of Sentiment Analysis Accuracy B...

引用

2025 Asia-Europe Conference on Cybersecurity, Internet of Things and Soft computing, CITSC 2025

作者： Ni, Jinyang School of Foreign Languages and Literature Shandong University Shandong Jinan China

ISBN: (纸本)9798331504205

The popularity of the internet has led to an explosion of users and consequent comment data on social media platforms such as X, once known as twitter. By analyzing the textual sentiment of such a large amount of public data, it is possible to count people's attitudes toward an event or a product. Such a visual survey can be useful for further monitoring of trends in public opinion or for updating products in accordance with the prevailing opinion. This study is dedicated to comparing the performance of different machine learning models in the setting of a shared dataset. The tweets of corresponding keywords are collected through open-source application programming interface (API) tools to form a shared dataset for this research. The dataset is used to train the Naive Bayes (NB), Artificial Neural Network (ANN) and Logistic Regression (LR) models respectively and the results of the model training are visualized through a confusion matrix. The model performance is then assessed by comparing the values of accuracy, precision, recall, and F1-score. It has been demonstrated that the LR model is better at dealing with the type of classification problem of sentiment analysis, which provides a referenceable suggestion for future directions. © 2025 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：