Grid Middleware and Services

Large-scale cooperation support for learners becomes even more important, when e-Learning is implemented in scalable, open, dynamic and heterogeneous environment. We have designed grid architecture and implemented grid middleware and higher CSCW services for establishing collaborative platform for e-Learning. A Learning Assessment Grid, abbreviated as LAGrid, is built on top of these services. This paper introduces the layered LAGrid architecture and presents how to implement grid middleware to support collaborative platform for e-Learning using Web services technologies. The work shows service-oriented grid middleware layer provides a good middleware platform for collaborative applications in large-scale cross-organization environment.

Grid middleware provides a way to integrate computational and storage resouces for supporting large-scale applications that span across multiple domains. Implicitly, Grid middlware eliminates the interoperability obstacle between different resources. However, with the emerging of a bunch of Grid middlewares, to provide interoperability between Grid middlewares themselves is an important challenge in production Grid infrasturtures. Web Service technologies (specifically, Simple Object Access Protocol) have been adopted in most of the Grid middlewares as the XML messaging protocol for the interoperability in the application layer. For other layers, standard protocols are also adopted for interoperability, e.g., HTTP is utilized as service transport protocol. On the other hand, security is a key issue that needs to be taken into account on each layer, for instance, WS-Security (Web Service Security) is considered as an augment on SOAP protocol for applying security to Web Services; GSI (Globus Security Infrastructure) is considered as an protocol for applying security to transport layer. We present the design consideration and implementation about how to provide flexible support for security protocols in the Advanced Resource Connector(ARC) Grid middleware, and this way clients or/and services developed in ARC middleware can easily interoperate with service/client developed in other middlewares, such as gLite and Globus Toolkit. Also, a flexible authorization framework is presented that can secure the Grid services with configurable authorization modules, as well as a variety of authorization policies.

Interactive Problem Solving Environments (PSEs) offer an integrated approach for constructing and running complex systems, such as distributed simulation systems. New distributed infrastructures, like the Grid, support the access to a large variety of core services and resources that can be used by interactive PSEs in a secure environment.We are experimenting with Grid access for interactive PSEs built on top of the High Level Architecture (HLA), a middleware for interactive simulations. Our current approach is such that once a PSE simulation has been executed in the framework, mechanisms from both HLA and Grid middleware are used to broker resources, for job submission services, performance monitoring services, and security services for efficient and transparent execution.We are experimenting with the Web-based Open Grid Services Architecture (OGSA) for HLA RTI Federate registration and discovery, as well as for data transmission. We have found that Grid Services give the possibility to allow for dynamic modification capabilities of HLA Federates, though the dynamic discovery of Federates and the use of Service Data (metadata) for service introspection is not trivial. Also, we have found that opening many data transmission channels between HLA Federates to one destination affects the number of connections you can make with other destinations.KeywordsPSEinteractiveGridOGSAHLA

To be effectively adopted in different application domains, next‐generation Grids need to address different issues such as: an increasing complexity and distribution of applications; different goals, skills and habits of Grid users; availability of different programming and deployment models; heterogeneous capabilities and performances of access networks and devices. Moreover, scientific and commercial applications, as well as Grid middleware, will increasingly produce an overwhelming quantity of application and usage data. Although the ongoing convergence between Grids, Web Services, and the Semantic Web constitutes a milestone towards a service‐oriented Grid architecture, which has the potential to face important issues such as application programming and business modelling, many other issues need research and development efforts. The great availability of data and information at the different layers of Grids, the maturity of data exploration techniques able to extract and synthesize knowledge, such as data mining, text summarization, semantic modelling, and knowledge management, and the demand for intelligent services in different phases of application life cycle are the driving forces towards novel knowledge‐based Grid services. Guided by those considerations, the paper first introduces main requirements of next‐generation Grids and then describes some representative knowledge‐based Grid services for both applications support and system management. Simple cases study showing how such services could be employed are discussed. Copyright © 2005 John Wiley & Sons, Ltd.

Grid technology has received wide acceptance in fields ranging from industry to academia. Its capability to integrate heterogeneous and distributed computing resources to provide transparent access has been leveraged in different areas of computing needs including healthcare provisioning. The global effort focuses on exploiting the grid infrastructure opportunities in the enhancement of health related activities. This paper reviewed few of the existing healthgrids in the context of applying existing grid infrastructures and middleware in providing specific medical services. The focus is on the services provided using grid services and infrastructures, the middleware forming the bases of interoperability between the services as well as service client specific applications. The contributions of this paper are to highlight and identify open problems, challenges within the healthgrid that can impede the goal of pervasiveness healthcare provisioning.

Secure cooperative access control on grid

The access to Grid resources depends on rules defined by the administrators of the physical organizations and of the Grid middleware. This approach does not require support for access control in the middleware, but since changes in the access control policy of the Virtual Organization imply the involvement of one or more administrators, it lacks the flexibility needed in a several application scenarios. In this paper we propose a group-based access control model for Grid environments that increases the flexibility of the access control model offered by state-of-the-art Grid platforms without requiring changes in the middleware. The approach is based on collaboration among Grid users and allows them to exchange access permissions to Virtual Resources without the intervention administrators. We show that our solution can be defined on top of the access control mechanisms offered by state-of-the-art Grid middleware and illustrate how the proposed model can be implemented as a service in a service-oriented Grid environment.

최근 증가하는 협업 연구는 원격 의료 데이터 공유 및 외부 기관의 데이터 접근을 요구하고 있다. 본 논문에서는 융통성 있고 효율적인 의료 데이터 관리를 위해 웹서비스를 이용하여 그리드와 PACS 간의 상호운용성 프레임워크를 제안하고 구현하였다. DICOM 표준은 PACS와 이미지 데이터베이스 사이에 의료 이미지 데이터의 교환과 전송을 정의한다. 그러나 병원들 간 의료 데이터 교환은 신뢰할 수 있는 기관들 사이에서만 제공된다. 더군다나 DICOM은 의료 데이터 관리 기능을 제공하지 않고 그리드 미들웨어는 DICOM 데이터에 접근 가능한 표준 툴킷을 제공하고 있지 않고 있다. 본 논문에서는 이러한 문제를 해결하기 위하여 PACS 간의 연동과 의료 데이터 관리를 가능하게 하는 웹 서비스기반의 그리드 서비스 중계자(Web Services-based Grid Service Mediator, WGSM)를 개발하였다. WGSM은 압축 중계자, GridFTP 중계자, RFT 중계자, MyProxy 중계자, MDS 중계자, RLS 중계자 등으로 구성된다. 제안한 웹서비스 기반의 프레임워크는 협업 환경에서 PACS 간 사용자 인증 및 안전한 데이터 접근을 제공한다. 특히, WGSM은 일반 사용자들이 그리드 미들웨어에 대한 어떠한 지식이 없어도 간단하고 효율적인 방법으로 원격 PACS를 접근할 수 있게 한다. Recently the increasing collaborative research requires the remote medical and clinical data sharing and access of external institutions. In this paper, an interoperability framework between Grid and PACS using Web services is proposed and implemented in order to provide flexible and efficient medical data management. The Digital Imaging and Communications in Medicine(DICOM) standard defines medical image data exchange and transfer between PACSs and image databases. However, medical data exchange between hospitals is limited within the trusted and static environments. Moreover, DICOM does not provide medical data management and the Grid middleware does not include standard toolkit to access DICOM data. To address this issue, a Web services-based Grid Service Mediator (WGSM) which provides PACS integration and medical image data management is developed. The WGSM consists of several service mediators such as compress mediator, GridFTP mediator, RFT mediator, MyProxy mediator, MDS mediator, and RLS mediator and others. The proposed Web services-based framework provides user authentication and secure data access between PACSs in collaborative environments. In particular, the WGSM allows ordinary users to access remote PACS data in a simple and efficient manner without any the knowledge about underlying Grid middleware.

In this paper, we present a solution to the probl of dynamically deploying grid service factories onto computing nodes running an implentation of the Open Grid Services Infrastructure (OGSI). By providing a non-intrusive Hot Deployment Service (HDS), we extend the service-oriented grid computing paradigm, as it is defined by the Open Grid Services Architecture (OGSA), to provide a more dynamic ad hoc grid environment. Service-oriented grid middleware utilizing the HDS enables organizations or interorganizational communities to form an ad hoc grid to harness unused and scattered resources of an existing IT-infrastructure. The availability of the HDS also improves the capabilities to manage existing grid systs based on the Globus Toolkit 3, which is a vital requirent for the adoption of service-oriented grid systs in production environments.

This paper reports on the design, implementation and performance evaluation of a suite of GridRPC programming middleware called Ninf-G Version 2 (Ninf-G2). Ninf-G2 is a reference implementation of the GridRPC API, a proposed GGF standard. Ninf-G2 has been designed so that it provides 1) high performance in a large-scale computational Grid, 2) the rich functionalities which are required to adapt to compensate for the heterogeneity and unreliability of a Grid environment, and 3) an API which supports easy development and execution of Grid applications. Ninf-G2 is implemented to work with basic Grid services, such as GSI, GRAM, and MDS in the Globus Toolkit version 2. The performance of Ninf-G2 was evaluated using a weather forecasting system which was developed using Ninf-G2. The experimental results indicate that high performance can be attained even in relatively fine-grained task-parallel applications on hundreds of processors in a Grid environment.

PSG: Peer-to-Peer semantic grid framework architecture

This paper presents an approach to development of interactive Grid applications used in the application development platform (Appea), currently under development. The presented environment enables smooth and reliable development of such applications, integrating various types of Web and Grid services as well as data sources spread over a distributed infrastructure. However, the users to which Appea caters require that such applications be capable of interactivity and require human actors to interact with them. The addition of interactivity requirements has not thus far been sufficiently addressed in the Grid domain and indeed many popular Grid middleware platforms do not take sufficient notice of this problem. Therefore, the aim of this paper is to describe how such aspects of user interactivity are supported by the Appea platform and how they correspond to the needs of end users of Grid applications. The paper presents the motivation behind the proposed solution, a conceptual design of Appea as well as its basic architecture. In addition, a use case is presented, showing how the solution can be applied to a real-life workflow.

Motivations. In the last decade many projects related to grids have been carried out in Europe at both national and international levels with an important economic contribution by the European Commission.The grid middleware developed within these projects has been deployed into the European grid infrastructure (EGI) made by more than 350 sites all over Europe. The Italian Grid Infrastructure (IGI) is part of EGI and is one of the most important and widest national grid infrastructures in Europe since it provides about 33000 CPU cores, 17 PB of disk space and 9 PB of tape capacity spread over more than 50 sites. Although the grid infrastructure has been initially built according to the needs of a few scientific communities (high energy physics, earth observation and Bioinformatics among the others), it has been gradually evolving in order to provide grid services to a wider and wider user base. Several scientific communities are observing that as the instruments become more and more powerful the need for storage and computing is increasing day by day. This will also increase the number of users that could benefit from a geographical distributed computing grid infrastructure. Methods. In order to support new communities (users and resource providers), various activities have been started. Training has been considered very important, so tutorial for users and grid administrators are regularly organized. In addition great effort has been devoted to understanding the user needs, by defining appropriate use cases, and to supporting the user communities to port their applications on the grid environment (the so-called "gridification"). An important effort is also spent in developing new tools that could make the interaction between the final users and the grid as easy as possible. In particular web tools to submit jobs to the grid infrastructure have been deployed and used by some bioinformatics communities. In order to address the needs of users relying on high-level tools like Workflow managers (e.g. Taverna and other similar tools), a front-end web service has been developed. This web interface could be used as a bridge towards the EGI/IGI grid infrastructure. The IGI community is also providing a service that allows the exploitation of Relational Databases over the grid infrastructure, assuring a high level of security and privacy. Results. The usage of the standard EGI/IGI resources and services, together with the high level services that IGI is providing on top of the grid, has provided the end users with the capability of carrying out their high demanding computing activities in an easy and reliable way. In the past years, indeed, IGI has supported several bioinformatics communities to "gridify" many different applications such as: ASPic, PAML, MrBayes, CSTGrid, DNAfan, BLAST, BayeSSC, FT-Comar, Muscle, Gene Ontology DB analysis, ABCtoolbox, EMBOSS, Bowtie, SAMtools, Illumina Solexa data processing, AmpliconNoise, BioPython, HMMER. As a result the CPU consumed by various Italian bioinformatics groups on the IGI grid infrastructure has exceeded the 10 years in a few days of activity, thus hugely reducing the overall time needed for the execution of the jobs. Thanks to IGI the bioinformatics users have carried out their analysis in an easy and transparent way, both through simple web interfaces and through complex WorkFlow managers, reducing the time needed to get the results of about 2-3 orders of magnitude. The IGI infrastructure has also been exploited by the Computational Biology group of the Bologna University, in the frame of DUCK (Distributed Unified Computing for Knowledge, a collaboration between multidisciplinary Academic and Research Institutions located in the Emilia Romagna region), to run protein annotation application based on the Bologna Annotation Resource (BAR) method, and to perform massively parallel genome sequencing including about 18 millions of protein sequences. More than 150 computing nodes in the IGI grid infrastructure have been used, successfully dropping the computational times if compared to the computing resources of the local cluster available to the group. The protein annotation application reached a dropping factor of 120, i. e. the computation was performed in few weeks instead of years. Data management facilities offered by the Grid were also exploited to easily handle input and output files. To provide an easy-to-use service to the user communities IGI is developing a web portal that will hide the complexity of the authentication/authorization mechanisms and will also integrate the computing frameworks needed by the different user communities. A prototype of this portal can be easily set up for the bioinformatics community. Availability http://www.italiangrid.it/

The aim of the EU funded EGEE (Enabling Grids for E-sciencE) project is to build on recent advances in Grid technology and develop a service Grid for scientific applications. This paper provides an overview the EGEE project goals and structure and explains the work being performed in the areas of grid operations, middleware re-engineering and application deployment and support. The relation of EGEE to digital libraries is explained and examples of how the project infrastructure can be used in the field are highlighted.

The Grid Component Model (GCM) is becoming a promising development platform for flexible and adaptable grid applications. Recently, a set of mediator components has been proposed for providing a uniform and integrated platform to access grid middleware, services, and resources from an application. In this paper, we present our experiences with building such mediator components using GCM, focusing on two functionality areas. First, we show how application adaption support can be realized viamediator components, based on a set of component controllers through which the application components can be adapted and steered. Second, we show how a service and resource abstraction layer can be controlled at runtime from the mediator components.