检索结果-内蒙古大学图书馆

19th international symposium on parallel and distributed computing (ISPDC)

作者： Alourani, Abdullah Kshemkalyani, Ajay D. Univ Illinois Dept Comp Sci Chicago IL 60607 USA Majmaah Univ Dept Comp Sci & Informat Al Majmaah 11952 Saudi Arabia

ISBN: (数字)9781728189468

ISBN: (纸本)9781728189468

Cloud computing offers a variable-cost payment scheme that allows cloud customers to specify the price they are willing to pay for renting spot instances to run their applications at much lower costs than fixed payment schemes, and depending on the varying demand from cloud customers, cloud platforms could revoke spot instances at any time. To alleviate the effect of spot instance revocations, applications often employ different fault-tolerance mechanisms to minimize or even eliminate the lost work for each spot instance revocation. However, these fault-tolerance mechanisms incur additional overhead related to application completion time and deployment cost. We propose a novel cloud market-based approach that leverages cloud spot market features to provision spot instances without employing fault-tolerance mechanisms to reduce the deployment cost and completion time of applications. We evaluate our approach in simulations and use Amazon spot instances that contain jobs in Docker containers and realistic price traces from EC2 markets. Our simulation results show that our approach reduces the deployment cost and completion time compared to approaches based on faulttolerance mechanisms.

关键词： cloud computing spot instances fault-tolerance mechanisms cloud spot market features cloud-based applications payment schemes spot instance revocations

来源：评论

学校读者我要写书评

暂无评论

QCCC 2022 Keynote Talk: Hybrid Quantum / Classical Algorithms for Machine Learning

QCCC 2022 Keynote Talk: Hybrid Quantum / Classical Algorithm...

引用

IEEE international symposium on parallel and distributed Processing Workshops and Phd Forum (IPDPSW)

作者： Nathan Wiebe University of Toronto

ISBN: (数字)9781665497473

ISBN: (纸本)9781665497480

In this talk I will provide a new approach to quantum machine learning that involves using classical machine learning to learn a representation for a dataset that can be embedded in a quantum computer. We will then consider applying this strategy to train a generative model for groundstates of chemistry Hamiltonians that will allow us to predict groundstates given data through a classically learnt representation that converts nuclear positions into weights for a quantum neural network that generates the state. This work shows that quantum / classical Hybrid methods can be a powerful way to learn how to generate groundstates and potentially even give a cheaper alternative to approximate groundstate preparation than phase estimation provides in some settings.

关键词： Quantum computing Computational modeling Phase estimation distributed processing Conferences Biological system modeling Thermodynamics

来源：评论

学校读者我要写书评

暂无评论

DFMan: A Graph-based Optimization of Dataflow Scheduling on High-Performance computing Systems

DFMan: A Graph-based Optimization of Dataflow Scheduling on ...

引用

international symposium on parallel and distributed Processing (IPDPS)

作者： Fahim Chowdhury Francesco Di Natale Adam Moody Kathryn Mohror Weikuan Yu Florida State University Lawrence Livermore National Laboratory

Scientific research and development campaigns are materialized by workflows of applications executing on high-performance computing (HPC) systems. These applications con-sist of tasks that can have inter- or intra-application flows of data to achieve the research goals successfully. These dataflows create dependencies among the tasks and cause resource con-tention on shared storage systems, thus limiting the aggregated I/O bandwidth achieved by the workflow. However, these I/O performance issues are often solved by tedious and manual efforts that demand holistic knowledge about the data dependencies in the workflow and the information about the infrastructure being utilized. Taking this into consideration, we design DFMan, a graph-based dataflow management and optimization framework for maximizing I/O bandwidth by leveraging the powerful storage stack on HPC systems to manage data sharing optimally among the tasks in the workflows. In particular, we devise a graph-based optimization algorithm that can leverage an intuitive graph representation of dataflow- and system-related information, and automatically carry out co-scheduling of task and data placement. According to our experiments, DFMan optimizes a wide variety of scientific workflows such as Hurricane 3D on Cloud Model 1 (CM1), Montage Carina Nebula (NGC3372), and an emulated dataflow kernel of the Multiscale Machine-learned Modeling Infrastructure (MuMMI I/O) on the Lassen supercomputer, and improves their aggregated I/O bandwidth by up to 5.42 x, 2.12 x and 1.29 x, respectively, compared to the baseline bandwidth.

关键词： Solid modeling Three-dimensional displays Processor scheduling High performance computing Bandwidth Manuals Supercomputers

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Data Transfer API for Rank Level Approximate computing on HPC Systems

Performance Evaluation of Data Transfer API for Rank Level A...

引用

IEEE international symposium on parallel and distributed Processing Workshops and Phd Forum (IPDPSW)

作者： Yoshiyuki Morie Yasutaka Wada Ryohei Kobayashi Ryuichi Sakamoto Faculty of Fukuoka Medical Technology Teikyo University Fukuoka JAPAN School of Information Science Meisei University Tokyo JAPAN Center for Computational Sciences University of Tsukuba Ibaraki JAPAN Global Scientific Information and Computing Center Tokyo Institute of Technology Tokyo JAPAN

ISBN: (数字)9781665497473

ISBN: (纸本)9781665497480

Approximate computing (AC) has attracted much attention to optimize tradeoffs among performance, power con-sumption, and computation results accuracy by adjusting data precision in applications. Even on HPC systems, AC is demanded to maximize performance under the limited power budget and hardware resources. To apply AC for HPC applications, we need to consider the character of each MPI rank in an application and optimize it with its appropriate data precision. However, we also need to perform data transfer while converting the precision of the target data. This paper proposes data pack/unpack APIs, which are applicable for standard MPI programs for HPC systems, for converting the data precision of the target data, and shows its performance evaluation. We can express data transfer among ranks with different data precision with the proposed APIs. The performance evaluation reveals the break-even point to apply AC for HPC applications from the perspective of data transfer volume.

关键词： Performance evaluation distributed processing Conferences Approximate computing Data transfer Hardware Standards

来源：评论

学校读者我要写书评

暂无评论

Towards a Cloud Native Big Data Platform using MiCADO 19

Towards a Cloud Native Big Data Platform using MiCADO

引用

19th international symposium on parallel and distributed computing (ISPDC)

作者： Mosa, Abdelkhalik Kiss, Tamas Pierantoni, Gabriele DesLauriers, James Kagialis, Dimitrios Terstyanszky, Gabor Univ Westminster Ctr Parallel Comp London England

ISBN: (纸本)9781728189468

In the big data era, creating self-managing scalable platforms for running big data applications is a fundamental task. Such self-managing and self-healing platforms involve a proper reaction to hardware (e.g., cluster nodes) and software (e.g., big data tools) failures, besides a dynamic resizing of the allocated resources based on overload and underload situations and scaling policies. The distributed and stateful nature of big data platforms (e.g., Hadoop-based cluster) makes the management of these platforms a challenging task. This paper aims to design and implement a scalable cloud native Hadoop-based big data platform using MiCADO, an open-source, and a highly customisable multi-cloud orchestration and auto-scaling framework for Docker containers, orchestrated by Kubernetes. The proposed MiCADO-based big data platform automates the deployment and enables an automatic horizontal scaling (in and out) of the underlying cloud infrastructure. The empirical evaluation of the MiCADO-based big data platform demonstrates how easy, efficient, and fast it is to deploy and undeploy Hadoop clusters of different sizes. Additionally, it shows how the platform can automatically be scaled based on user-defined policies (such as CPU-based scaling).

关键词： Hadoop big data platform MiCADO Docker Containers Container-orchestration cloud native scaling big data infrastructure

来源：评论

学校读者我要写书评

暂无评论

Batch-Schedule-Execute: On Optimizing Concurrent Deterministic Scheduling for Blockchains

Batch-Schedule-Execute: On Optimizing Concurrent Determinist...

引用

Reliable distributed Systems

作者： Yaron Hay Roy Friedman Technion Israel

ISBN: (数字)9798331530037

ISBN: (纸本)9798331530044

Executing smart contracts is a compute and storage-intensive task, which currently dominates modern blockchain's performance. Given that computers are becoming increasingly multicore, concurrency is an attractive approach to improve programs' execution runtime. A unique challenge of blockchains is that all replicas (miners or validators) must execute all smart contracts in the same logical order to maintain the semantics of State Machine Replication (SMR). In this work, we study the maximal level of parallelism attainable when focusing on the conflict graph between transactions packaged in the same block. This exposes a performance vulnerability that block creators may exploit against existing blockchain concurrency solutions, which rely on a total ordering phase for maintaining consistency amongst all replicas. To facilitate the formal aspects of our study, we develop a novel generic framework for Active State Machine Replication (ASMR) that is strictly serializable. We introduce the concept of graph scheduling and the definition of the minimal latency scheduling problem, which we prove to be NP-hard. We show that the restricted version of this problem for homogeneous transactions is equivalent to the classic Graph Vertex Coloring Problem, yet show that the heterogeneous case is more complex. We discuss the practical implications of these results.

关键词： Concurrent computing Computers Runtime Multicore processing Smart contracts Semantics Focusing parallel processing Blockchains Reliability

来源：评论

学校读者我要写书评

暂无评论

distributed and Fault-Tolerant Construction of Low Stretch Spanning Tree 19

Distributed and Fault-Tolerant Construction of Low Stretch S...

引用

19th international symposium on parallel and distributed computing (ISPDC)

作者： Gurjar, Aishwarya Peri, Sathya Sengupta, Sinchan Indian Inst Technol Dept Comp Sci & Engn Hyderabad India

ISBN: (纸本)9781728189468

Spanning trees are widely used as a communication backbone over some given infrastructure and help network designers achieve a low-cost communication overhead. Spanning trees are generally designed, keeping in mind some optimizing metric (most general being sum of edge weights in a Minimum Spanning Tree) with respect to the underlying graph. For applications that require preserving shortest path distances between nodes of the weighted underlying graph in the abstracted spanning tree, we look to minimize a parameter known as stretch. Stretch is defined as the ratio of the distance between two nodes in the tree to its shortest path distance in the communication graph. To make spanning-tree constructions resilient to edge failures in an error-prone environment, we consider what is called the All Best Swap Edges (ABSE) problem. Since every edge in a tree is a bridge edge, a single edge failure disconnects the tree into two connected components. In the ABSE problem, for each edge e in the spanning tree, we compute a swap edge f corresponding to e, that is activated when e fails. f helps to restore the communication in the tree by connecting the disconnected components. In this paper, we give a novel distributed algorithm to efficiently construct a low average stretch spanning tree and make it robust against edge failures by finding a swap edge for every edge in the constructed tree. This is the first known deterministic distributed algorithm for constructing a low stretch tree that is also edge fault-tolerant. The distributed ABSE computation in our case equals the state-of-the-art running time of O( h) rounds, where h is the height of the tree.

关键词： Spanning tree Stretch Swap edges distributed algorithms

来源：评论

学校读者我要写书评

暂无评论

P-ckpt: Coordinated Prioritized Checkpointing

P-ckpt: Coordinated Prioritized Checkpointing

引用

international symposium on parallel and distributed Processing (IPDPS)

作者： Subhendu Behera Lipeng Wan Frank Mueller Matthew Wolf Scott Klasky North Carolina State University Oak Ridge National Laboratory

Good prediction accuracy and adequate lead time to failure are key to the success of failure-aware Check-point/Restart (C/R) models on current and future large-scale High-Performance computing (HPC) systems. This paper develops a novel checkpointing technique, called p-ckpt, that aims to maintain the performance efficiency of failure-aware C/R models even when failures are predicted with a small lead time. The p-ckpt technique is developed for HPC systems with multi-level memory systems to prioritize checkpoints from vulnerable nodes (nodes with predicted failure) in the event of failure prediction. It applies coordination among the nodes within an application so that vulnerable nodes' checkpoint data is stored to the parallel File System (PFS) first by assigning priorities based on the lead time to failure. Vulnerable nodes thus have low-latency access on the critical path to the PFS before any failure happens. Further, we create the hybrid p-ckpt model by integrating Live Migration (LM) because of its cost-effectiveness and to reduce checkpoint frequency. Our hybrid p-ckpt C/R model considers prediction lead time and checkpoint latency to the PFS to decide on a feasible proactive action such as p-ckpt and LM. Simulations of six real-world applications for the Summit supercomputer show a ≈53-65% reduction in overhead due to the hybrid p-ckpt model compared to a ≈31-61% reduction in a state-of-the-art solution. We assess our C/R models against multiple failure distributions and consider lead time variability and failure prediction accuracy. Based on this evaluation and assessment, we discuss the trade-offs of using these models and their impact on application overhead.

关键词： Checkpointing Fault tolerance distributed processing File systems Computational modeling High performance computing Fault tolerant systems

来源：评论

学校读者我要写书评

暂无评论

Optimistic scheduling with service guarantees

引用

JOURNAL OF parallel AND distributed computing 2020年 135卷 246-258页

作者： Kambatla, Karthik Yarlagadda, Vamsee Goiri, Inigo Grama, Ananth Facebook Inc 1 Hacker Way Menlo Pk CA 94025 USA Cloudera Inc 395 Page Mill Rd Palo Alto CA USA Microsoft Res 14865 NE 36th St Redmond WA USA Purdue Univ Dept CS 305 N Univ Ave W Lafayette IN 47907 USA

Data centers form the core of current information and commercial enterprise. At current scale, any improvement in data center resource utilization leads to substantial savings. We focus on the problem of scheduling jobs in distributed execution environments to improve resource utilization. Cluster schedulers like YARN and Mesos base their scheduling decisions on resource requirements provided by end users. It is hard for end-users to predict the exact amount of resources required for a task/job, especially since resource utilization can vary significantly over time and across tasks. In practice, users make highly conservative estimates of peak utilization across all tasks of a job to ensure job completion, leading to resource fragmentation and severe under utilization in production clusters. We present UBIS, a utilization-aware approach to cluster scheduling, to address resource fragmentation, and to improve cluster utilization and job throughput. UBIS considers actual usage of running tasks and schedules opportunistic work on under-utilized nodes. UBIS monitors resource usage on these nodes and preempts opportunistic containers in the event this over-subscription becomes untenable. In doing so, UBIS effectively utilizes wasted resources, while minimizing adverse effects on regularly scheduled tasks. Our implementation of UBIS on YARN demonstrates improvements of up to 30% in makespan for a representative workload and 25% in individual job durations. (C) 2019 Elsevier Inc. All rights reserved.

关键词： Cluster scheduling distributed computing

来源：评论

学校读者我要写书评

暂无评论

Generalized Flow-Graph Programming Using Template Task-Graphs: Initial Implementation and Assessment

Generalized Flow-Graph Programming Using Template Task-Graph...

引用

international symposium on parallel and distributed Processing (IPDPS)

作者： Joseph Schuchart Poornima Nookala Mohammad Mahdi Javanmard Thomas Herault Edward F. Valeev George Bosilca Robert J. Harrison Innovative Computing Laboratory The University of Tennessee Knoxville TN USA Institute for Advanced Computational Science Stony Brook University Stony Brook NY USA Meta Platforms Inc New York NY USA Department of Chemistry Virginia Polytechnic Institute and State University Blacksburg VA USA

We present and evaluate TTG, a novel programming model and its C++ implementation that by marrying the ideas of control and data flowgraph programming supports compact specification and efficient distributed execution of dynamic and irregular applications. Programming interfaces that support task-based execution often only support shared memory parallel environments; a few support distributed memory environments, either by discovering the entire DAG of tasks on all processes, or by introducing explicit communications. The first approach limits scalability, while the second increases the complexity of programming. We demonstrate how TTG can address these issues without sacrificing scalability or programmability by providing higher-level abstractions than conventionally provided by task-centric programming systems, without impeding the ability of these runtimes to manage task creation and execution as well as data and resource management efficiently. TTG supports distributed memory execution over 2 different task runtimes, PaRSEC and MADNESS. Performance of four paradigmatic applications (in graph analytics, dense and block-sparse linear algebra, and numerical integrodifferential calculus) with various degrees of irregularity implemented in TTG is illustrated on large distributed-memory platforms and compared to the state-of-the-art implementations.

关键词： Runtime Processor scheduling Scalability distributed databases C++ languages Programming Data models

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：