高性能计算环境通用计算平台

随着高性能计算环境的持续运行,用户的计算需求快速增长,对资源的需求不断上升.环境急需扩充资源以提供更强大的计算能力,满足不同应用领域的计算需求,同时也对环境扩展性、易用性和可靠性提出更高要求.高性能计算环境通用平台从底层设计到上层页面实现了全新的蜕变,从用户角度出发设计了相关的功能模块,为用户提供更加优质的服务.本文围绕国家高性能计算环境通用计算平台展开叙述,重点介绍平台的整体结构、采用的关键技术、相关模块以及实现过程和最终效果.通过部署测试表明通用计算平台可满足现有用户的需求并将进一步扩大高性能计算环境的用户群,推动高性能计算在更多领域的发展,提高国家高性能计算环境影响力.     

高性能计算服务环境应用编程接口

高性能计算服务环境主要面向用户、科研团队提供高性能计算服务. 随着环境接入的超算中心以及应用社区和业务平台越来越多, 超算中心以及社区和业务平台的用户希望能够使用原有账号登录高性能计算环境使用资源. 高性能计算服务环境目前提供的应用编程接口仅支持通过LDAP认证的网格账号. 为使得应用社区和业务平台用户使用自己原有的登录方式认证通过后就可访问高性能计算服务环境, 我们重新设计开发了高性能计算服务环境应用编程接口. 本文着重介绍新版应用编程接口的结构与部署实现, 并通过用例来说明如何调用新版接口. 新版接口为社区和业务平台接入高性能计算环境提供了更方便且安全地支撑.     

地震勘探集成环境资源监控研究与实现

利用网格技术建设地震勘探集成环境高性能计算平台基础设施的过程中,监控系统不仅要为位于不同管理域、分布异构的高性能资源管理提供资源全局视图,方便对资源管理和控制,还要为分布环境下资源的选择、任务的调度与执行提供指导信息．本文提出了面向地震勘探领域的高性能应用网格监控系统体系结构,建立了监控信息数据模型,在GlobusToolkit信息服务组件的基础上扩展了信息采集功能和信息查询接口,并基于Websphere为高性能计算资源的信息查询与监控提供了资源实体的统一视图和可视化管理Portal．经测试,系统运行稳定,可扩展性强．     

面向材料科学研究的新型高性能计算平台系统

高性能计算集群作为保证国家科研开展的“基础设施”已上升为国家战略。高性能计算应用广泛,特别在材料科学研究方面,是必不可少的工具。当下为科学计算用户提供高质量的远程化、可视化、图形化的高性能计算平台成为当前高性能服务研究的突破口。本文提出基于材料科学研究的新型高性能计算平台系统,该系统基于Java语言开发,采用B/S架构为用户提供服务,实现主流材料科学研究的计算软件与平台的整合,设计友好的用户操作界面,提供方便的接入方式,并结合OpenPBS优化作业调度方法为平台用户提供优先级更高的计算需求,保证材料科研应用中更高效的计算资源。     

Lyra——多域环境下的分布资源管理系统

随着应用需求的不断增长,并行计算环境的规模越来越大,高性能计算环境由单个集群向多个集群扩展,由本地资源向远程资源扩展,所涉及的计算资源种类繁多,系统管理也更加复杂.本文介绍的Lyra系统是一个针对多域计算环境特点设计的大规模分布资源管理系统,具有资源模型适应性强、系统可扩展性好、资源调度策略灵活、对多域并行计算支持能力强的特点,可以为大规模并行计算和大规模信息处理系统提供资源监控管理支撑以及资源信息服务.本文介绍了LYRA系统的设计结构,分析了系统的关键技术特点,并给出了与相关系统的比较.     

基于资源目录技术的网格系统软件设计与实现

由于网格计算系统中地理分布着大量不同类型动态实体，资源、服务和计算的发现、特征化、监控变得非常困难，资源信息服务成为网格软件的关键技术之一。采用面向对象方法设计网格资源信息模型，使用目录服务技术建造网格系统软件中的资源信息服务，提供了资源信息发现、收集、存储和访问机制，介绍了国家高性能计算环境的基本结构和功能，详细描述了基于轻权目录访问协议（LDAP）的网格系统软件的设计和实现。     

基于高性能计算平台的地学集成模拟分析环境研究

随着计算科学在地学领域中广泛的应用以及地学各个学科模拟理论研究的发展,目前涌现出了一批功能各异的地学计算模型。地学工作者在科学研究中要快速便捷地熟悉和使用这些模型,迫切需要一个集成的模拟分析环境,即把硬件与软件结合,使用和服务结合,并关联相应的模型文献、驱动数据、处理工具、源代码等。在本文中我们基于高性能计算平台软硬件环境建设的工作经验和地学发展趋势,提出了基于高性能计算平台的地学集成模拟分析环境 GISE（GeoscienceIntegrated Simulation Environment）总体设计框架,并对该环境中核心部分进行了设计和详细介绍。GISE 将充分有效地集成各类模型资源,并将 HPC 服务和应用相结合,以期形成地学 e-Science 的核心信息平台。     

面向云服务的高性能计算柔性服务平台

随着云计算的深入应用,为用户提供高质量的远程高性能云服务成为当前研究的热点。文章提出了面向云服务的高性能计算柔性服务平台,该平台通过虚拟化技术、图形化和面向对象的技术等,设计和实现了全新的高性能计算环境,具有工具整合程度高、用户界面友好等特点,可为远程云服务提供必要支持。     

三层架构超级计算环境容错框架*

为确保建成的中国科学院“十一五”信息化重大专项超级计算环境提供稳定可靠的服务,提出三层架构超级计算环境的容错框架。对计算环境可靠性和计算节点可靠性两大部分,通过作业可靠性、服务可靠性和网格节点可靠性三个主要方面的可靠性研究,提出并实现了三层架构超级计算环境的可靠性解决方案。该框架重点解决了单点故障对环境的影响,确保单点故障发生后系统能够继续提供高可用的高性能计算服务。     

国家高性能计算环境事件流系统的设计

国家高性能计算环境是由中国众多国家级计算中心和高校的计算集群聚合而成的大型高性能计算环境,为国内研究人员提供优质计算资源。出于维护环境正常稳定运行的目的,环境管理人员需要获取环境内部所发生的各种事件信息,以确保及时迅速地对环境产生的问题进行处理。针对这种需求,设计了国家高性能计算环境事件流处理与分发系统,用于对环境各类事件进行收集和按类型分类,最终提供给对事件有需求的环境应用。在该系统中,事件工厂模块负责对环境的各种事件进行格式解析以及初步过滤和处理等加工工作,然后将加工过的事件封装为统一的接口格式对外发布。初步实现了事件流系统的各部分功能,将其部署到国家高性能计算环境中,并对该系统的事件处理延时进行测试。实验结果表明事件处理过程的延时很低,可以满足对事件时效性的要求。     

Mobile agent middleware for mobile computing

Mobile computing requires an advanced infrastructure that integrates suitable support protocols, mechanisms, and tools. This mobility middleware should dynamically reallocate and trace mobile users and terminals and permit communication and coordination of mobile entities. In addition, open and untrusted environments must overcome system heterogeneity and grant the appropriate security level. Solutions to these issues require compliance with standards to interoperate with different systems and legacy components and a reliable security infrastructure based on standard cryptographic mechanisms and tools. Many proposals suggest using mobile agent technology middleware to address these issues. A mobile agent moves entities in execution together with code and achieved state, making it possible to upgrade distributed computing environments without suspending service. We propose three mobile computing services: user virtual environment (UVE), mobile virtual terminal (MVT), and virtual resource management (VRM). UVE provides users with a uniform view of their working environments independent of current locations and specific terminals. MVT extends traditional terminal mobility by preserving the terminal execution state for restoration at new locations, including active processes and subscribed services. VRM permits mobile users and terminals to maintain access to resources and services by automatically requalifying the bindings and moving specific resources or services to permit load balancing and replication     

An efficient job management of computing service using integrated idle VM resources for high-performance computing based on OpenStack

In recent years, various studies on OpenStack-based high-performance computing have been conducted. OpenStack combines off-the-shelf physical computing devices and creates a resource pool of logical computing. The configuration of the logical computing resource pool provides computing infrastructure according to the user’s request and can be applied to the infrastructure as a service (laaS), which is a cloud computing service model. The OpenStack-based cloud computing can provide various computing services for users using a virtual machine (VM). However, intensive computing service requests from a large number of users during large-scale computing jobs may delay the job execution. Moreover, idle VM resources may occur and computing resources are wasted if users do not employ the cloud computing resources. To resolve the computing job delay and waste of computing resources, a variety of studies are required including computing task allocation, job scheduling, utilization of idle VM resource, and improvements in overall job’s execution speed according to the increase in computing service requests. Thus, this paper proposes an efficient job management of computing service (EJM-CS) by which idle VM resources are utilized in OpenStack and user’s computing services are processed in a distributed manner. EJM-CS logically integrates idle VM resources, which have different performances, for computing services. EJM-CS improves resource wastes by utilizing idle VM resources. EJM-CS takes multiple computing services rather than single computing service into consideration. EJM-CS determines the job execution order considering workloads and waiting time according to job priority of computing service requester and computing service type, thereby providing improved performance of overall job execution when computing service requests increase.

     

Ontology-based access control model for security policy reasoning in cloud computing

There are many security issues in cloud computing service environments, including virtualization, distributed big-data processing, serviceability, traffic management, application security, access control, authentication, and cryptography, among others. In particular, data access using various resources requires an authentication and access control model for integrated management and control in cloud computing environments. Cloud computing services are differentiated according to security policies because of differences in the permitted access right between service providers and users. RBAC (Role-based access control) and C-RBAC (Context-aware RBAC) models do not suggest effective and practical solutions for managers and users based on dynamic access control methods, suggesting a need for a new model of dynamic access control that can address the limitations of cloud computing characteristics. This paper proposes Onto-ACM (ontology-based access control model), a semantic analysis model that can address the difference in the permitted access control between service providers and users. The proposed model is a model of intelligent context-aware access for proactively applying the access level of resource access based on ontology reasoning and semantic analysis method.     

海云创新试验环境管控与服务系统总体设计

资源管控与服务系统是海云创新试验环境的神经中枢和纽带.本文提出创新试验环境管控与服务系统的总体架构、协议与标准,指导创新试验环境的设计与实现.在此基础上开展系统的框架和功能设计,研究资源管理控制技术、试验服务技术、测量及联邦技术,实现试验环境的动态集成、管理与实时监控、试验活动的自适应资源分配与运行调度、试验过程的多维多粒度、全程全息测量等,从而为试验用户提供可管可控可测的综合试验服务与开发环境,提供公共资源与服务,动态集成、演示和评估专项研发成果,交互式地支撑专项创新工作.     

Resource collaboration system based on dynamic user preference and context

In a ubiquitous computing environment, service composition and collaboration among heterogeneous resources are required, thus, an infrastructure that supports these requirements is an essential factor in seamless service delivery. In this environment, users hope to get a variety of customized services by using only an individual mobile device. But the resource of the mobile device has limitations such as tiny display screens, limited input, less powerful processors, and limited storage. Moreover each user situation is different and user preferences are also various. Therefore it is one of new issues to provide a customized service for a user through resource collaboration based on various user preference and situation. To solve this issue, this paper proposes a resource collaboration system which infers customized resources for composing a user required service and collaborative with selected resources. For our collaboration system, this paper proposes the method to infer resources based on the context and user preferences including dynamic change of the preference. This paper also shows a reasonable execution environment for the proposed system through the performance evaluation in server-client and peer-to-peer environments.     

Interoperable access framework for internet of things backhauled heterogeneous applications

Internet of Things (IoT) provides large-scale interoperable service delegations for heterogeneous users through pervasive access. The interconnection of different devices and services in an on-demand and pervasive manner provides seamless responses for different applications. However, a few interconnected devices are stuck into a resource-constraint problem due to limited access and unavailability. This article proposes an interoperable access framework (IAF) for pervasive application services (PAS) for providing robust end-user solutions. The proposed framework balances the accessibility and semantics and its responses for providing seamless interoperability support. In this framework, transfer learning for different limited and free-flow service accessibility is used. The teaching is recurrent for identifying the unavailability of the resources. The process of serving the requests is achieved through interoperable replacement services. It reduces the impact of resource-constraint in the IoT platform, wherein the semantics between unavailability and accessibility is analyzed flawlessly. The proposed framework's performance is verified using the metrics service delay, responses, unavailability, resource utilization, and successful access.     

A specification model for context-based collaborative applications

This paper presents a model for building context-based systems in pervasive computing environments from high level specifications. A pervasive computing environment is viewed as a collaboration space consisting of mobile users, system services, and sensors and resources embedded in the physical environment. The approach presented here is developed for building collaborative activities in which users and environment services cooperate towards some shared objectives and tasks. The specification model allows expression of policies related to context-based discovery and secure access of resources, and role-based interactions among users and environmental services. Using several examples we illustrate the capabilities of our specification model for expressing various kinds of context-based requirements for resource access and user interactions.     

Dynamic Service Composition in Pervasive Computing

Service-oriented architectures (SOAs) promise to provide transparency to resource access by exposing the resources available as services. SOAs have been employed within pervasive computing systems to provide essential support to user tasks by creating services representing the available resources. The mechanism of combining two or more basic services into a possibly complex service is known as service composition. Existing solutions to service composition employ a template-matching approach, where the user needs are expressed as a request template, and through composition, a system would identify services to populate the entities within the request template. However, with the dynamism involved in pervasive environments, the user needs have to be met by exploiting available resources, even when an exact match does not exist. In this paper, we present a novel service composition mechanism for pervasive computing. We employ the service-oriented middleware platform called pervasive information communities organization (PICO) to model and represent resources as services. The proposed service composition mechanism models services as directed attributed graphs, maintains a repository of service graphs, and dynamically combines multiple basic services into complex services. Further, we present a hierarchical overlay structure created among the devices to exploit the resource unevenness, resulting in the capability of providing essential service-related support to resource-poor devices. Results of extensive simulation studies are presented to illustrate the suitability of the proposed mechanism in meeting the challenges of pervasive computing user mobility, heterogeneity, and the uncertain nature of involved resources.     

A lightweight conditional privacy-preserving authentication and access control scheme for pervasive computing environments

In pervasive computing environments, the users can get access to the services from the service providers in a highly desirable way. But the security of the user's authentication is a challenging field. Pervasive computing environments must provide the service to only legitimate users. On the other hand, some users attempt to keep their anonymity without revealing their identities while using some privacy-related services such as location information, printing, buying shares, etc. In this paper, we propose a conditional privacy-preserving authentication and access control scheme for pervasive computing environments, called CPriauac. Compared with the previous schemes in the literature, registration servers and authentication servers in the proposed scheme need not maintain any sensitive verification tables. The management of public keys is easier. Furthermore, the anonymity of the user can be removed efficiently once the dispute happens. The proposed scheme provides user anonymity against outside and inside parties, mutual authentication, accountability and differentiated access control.