- Over the past few years, various international groups have initiated research in the area of parallel and distributed computing in order to provide scientists with new programming methodologies that are required by state-of-the-art scientific application domains. - The concept of the Grid is based on coordinated resource sharing and problem solving in dynamic multi-institutional virtual organizations [1]. - Researchers therefore have been developing basic and advanced services, and portals for these services, to facilitate the realization of such com- plex environments and to hide the complexity of the underlying infrastructure. - We then illustrate the creation of a GCE with the help of commodity technologies provided through the Java framework. - Next, we outline differences from other CoG Kits and provide an overview of ongoing research in the Java CoG Kit Project, which is part of the Globus Project.. - Figure 26.1 A Grid computing environment hides many of the complex interactions between the accessible services.. - The user is oblivious of the fact that a service may engage other services on his or her behalf.. - A Grid portal provides information about the status of the Grid resources and services. - Grid portals provide communities another advantage: they hide much of the complex logic to drive Grid-related services with simple interaction through. - All of these services are examples of the many possible services in a GCE and are based on the most elementary Grid services. - The origins of this working group can be traced back to the Desktop Access to Remote Resources organization that was later renamed to ComputingPortals.org and are spin-offs from the Java Grande Forum efforts [21].. - GCEs are usually developed by reusing a number of commodity technologies that are an integral part of the target environment. - This effort has made progress with the introduction of the Global Grid Forum and pioneering projects such as the Globus Project. - Because of the use of different commodity technologies as part of different application requirements, a variety of CoG Kits must be supported. - As good examples, we present the Java and the Python CoG Kits for the Globus Toolkit, known as Java CoG and pyGlobus, respectively. - However, it is important to recognize the different approaches the Java and the Python CoG Kit pursue.. - While the Python CoG Kit interfaces with the Globus Toolkit on an API-based level, the Java CoG Kit interfaces with Globus services on a protocol level. - Kit assumes the availability of precompiled Globus Toolkit libraries on the current host- ing system, while the Java CoG Kit is implemented in pure Java and does not rely on the C-based Globus Toolkit. - The first version of the Java CoG Kit was based on Java Native Interface (JNI) wrappers for the Globus Toolkit APIs.. - This approach, however, severely restricted the usage of the Java CoG Kit for developing pure Java clients and portals that are to be executed as part of browser applets. - The availability of the functionality of the Globus Toolkit in another language has proved valuable in providing portability and assurance of code quality through protocol compliance.. - Both the Python and Java CoG Kits provide additional value to Grids over and above a simple implementation of the Globus Toolkit APIs. - The use of the commodity technolo- gies such as object orientation, stream management, sophisticated exception, and event handling enhances the ability to provide the next generation of Grid services. - We now focus on the Java CoG Kit as an example CoG Kit, and illustrate how it can be used to effectively build components that can be reused in the implementation of a GCE.. - 26.4 OVERVIEW OF THE JAVA COG KIT. - We have depicted in Figure 26.3 a small subset of the Java technology that can be used to support various levels of the Grid architecture [1]. - The Java CoG Kit builds a bridge between existing Grid technologies and the Java framework while enabling each to use the other’s services to develop Grid services based on Java technology and to expose higher-level frameworks to the Grid community while providing interoperability [9]. - The Java CoG Kit provides convenient access to the functionality of the Grid through client side and a limited set of server-side classes and components.. - We are currently investigating these and other technologies for Grid computing as part of the Commodity Grid projects to prototype a new generation of Grid services.. - Grid services framework Java CoG Kit Objects Java framework. - Figure 26.3 The Java CoG Kit allows users to access Grid services from the Java framework and enables application and Grid developers to use a higher level of abstraction for developing new Grid services and GCEs.. - The Java CoG Kit is general enough to be used in the design of a variety of advanced Grid applications with different user requirements. - The Java CoG Kit integrates Java and Grid components and services within one toolkit, as a bag of services and components.. - The goal of the Java CoG Kit is to enable Grid developers to use much of the Globus Toolkit functionality and to have access to the numerous additional libraries and frameworks developed by the Java community, allowing network, Internet, enterprise, and peer-to-peer computing. - Since the Java CoG Kit strives to be only protocol compliant, it does not provide a simple one-to-one mapping between the C Globus Toolkit and Java CoG Kit API. - A data transfer compatible with a subset of the Globus Toolkit GridFTP [37] and/or GSIFTP [38].. - Additionally, the Java CoG Kit contains a set of command-line scripts that provide convenient access to Globus Toolkit-enabled production Grids from the client. - Besides these elementary Grid services and tools, several other features and services currently not provided by the C Globus Toolkit are included explicitly or implicitly within the Java CoG Kit.. - Java Web- start allows the easy distribution of the code as part of downloadable jar files that are installed locally on a machine through a browser or an application interface. - We have demonstrated the use of Webstart within the Java CoG Kit by installing sophisticated Graphical User Interface (GUI) applications on client machines. - Thus, our goal of developing collaborative scientific problem-solving environments and portals, based on the combined strength of the Java and the Grid technologies, is well substantiated by the Java CoG Kit. - In the past, we had proposed portal architectures similar to the one depicted in Figure 26.2, in which the Java CoG Kit is used as an elementary middleware to integrate Grid services within portals and applications. - We expect that advanced services will be integrated in future releases within the Java CoG Kit or as extension packages. - Additionally, it is possible to implement several core Grid services, currently provided as part of the C Globus Toolkit, in pure Java while exposing the service through the Web Services Framework proposed recently by W3C. - We have provided example programs using advanced GUI components in Java as part of the Java CoG Kit. - We hope that the community will contribute more components so that the usefulness of the Java CoG Kit will increase.. - Our current work is focused on the creation of an extended execution service and the integration of Web services in our CoG Kit efforts. - An important result from this prototyping has been the development of the ‘InfoGram’. - Currently, we are also exploring the use of the InfoGram Service as part of ‘Sporadic Grids’, which are computational Grids dealing with sporadically available resources such as a computer at a beamline or a computer donated for a short period of time to a compute cluster.. - Besides executing processes outside of the JVM, we have enhanced the security model for Grid computing while reusing Java’s security model to, for example, restrict access to machine resources and prevent Trojan programs.. - Despite all the advantages of the Web service technology, currently there are only limited Web service development environments, especially in languages other than Java. - The converter generates client- and server-side code for the target Web service using the standard Axis WSDL2Java converter, as well as the code for the actual implementation of the Web service using the XML-based mapping that has been provided.. - An example of the mapping, which has been used as part of the CMCS project [4], is shown in the Figure 26.4. - 26.6 ADVANCED COG KIT COMPONENTS. - Now that we have illustrated the usefulness of CoG Kits, using the example of the Java CoG Kit, we demonstrate how we use it to provide clients with access to advanced services. - The associated output of the remote job can be downloaded by clicking on the appropriate file descriptor in the monitoring component. - The specification of the icons and the associations to jobs and machines are represented in XML format. - Figure 26.5 A prototypical GUI component performing job management for the GCE using the Java CoG Kit.. - Figure 26.6 A prototypical GUI performing data management for the GCE using the Java CoG Kit.. - Figure 26.7 A prototypical component using the Java CoG Kit to perform workflow for the GCE.. - The user community served by the Java CoG Kit is quite diverse. - The Java CoG Kit allows. - middleware developers to create new middleware components that depend on the Java CoG Kit;. - A subset of projects currently using the Java CoG Kit for accessing Grid functionality includes the following:. - CoGBox [50] provides a simple GUI for much of the client-side functionality such as file transfer and job submission.. - The NASA IPG LaunchPad [55] uses the GPDK based on the Java CoG Kit. - The Astrophysics Simulation Code Portal (ASC Portal) [57] is building a computa- tional collaboratory to bring the numerical treatment of the Einstein Theory of General Relativity to astrophysical problems.. - The CoG Kit provides access to the Grid.. - The Java CORBA CoG Kit [61] provides a simple Grid domain that can be accessed from CORBA clients. - uses the Java CoG Kit to interface with Globus.. - Additionally, work is currently performed as part of the Globus Project to provide a reference implementation of the Open Grid Services Architecture (OGSA) proposed through the Global Grid Forum. - The current technology preview uses the Java CoG Kit’s GSI security implementation and a modified version of the Java CoG Kit’s GRAM. - Java CoG Kit NCSA. - Figure 26.8 The Java CoG Kit builds a solid foundation for developing Grid applications based on the ability to combine Grid and Web technologies.. - Java Java CoG Kit Protocol-based www.globus.org/cog. - CoG Kit. - The role of the Java CoG Kit for some of these projects is depicted in Figure 26.8.. - We encourage the users to notify us of additional projects using CoG Kits, so that we can receive feedback about the requirements of the community. - As part of the Java and Python Commodity Grid project, we provide an elementary set of classes that allow the Java and Python programmers to access basic Grid services, as well as enhanced services suitable for the definition of desktop problem-solving environments.. - Additionally, we provided the Globus Toolkit with an independent set of client tools that was able to increase the code quality of the C Globus Toolkit and the productivity of the end user.. - Our future work will involve the integration of more advanced services into the Java CoG Kit and the creation of other CoG Kits and the integration of Web services tech- nologies. - We emphasize that a CoG Kit provides more than just an API to existing Grid services. - Indeed, it brings the modalities and the unique strength of the appropriate commodity technology to the Grid as the Grid brings its unique strengths to the commodity users. - The Java CoG Kit closely monitors the development within the Globus Project to ensure that interoperability is maintained. - The CoG Kit development team continues to keep track of projects that use the Java CoG Kit and documents the requirements of the community, in order to feed this information back to the Globus development team and to develop new features within the Java CoG Kit. - For up-to-date release notes, readers should refer to the Web page at http://www.globus.org/cog, where the Java CoG kit is available for download.. - This work is supported by the Mathematical, Information, and Computational Science Division subprogram of the Office of Advanced Scientific Computing Research, US Department of Energy, under Contract W-31-109-Eng-38. - Fox, Dennis Gannon, Shawn Hampton, Manish Parashar, Snigdha Verma, Mary Thomas, and Jay Alameda for the valuable discussions during the course of the ongoing CoG Kit development. - We thank Nell Rehn, Peter Lane, Pawel Plaszczak, Mike Russell, Jason Novotny, Ravi Madduri, Benjamin Temko, Shamsuddin Ladha, Beulah Alunkal, and all others who have contributed to the Java CoG Kit. - This work would not have been possible without the help of the Globus Project team. - The Java CoG Kit is developed at Argonne National Laboratory as part of the Globus Project.. - (2001) The anatomy of the grid: enabling scalable virtual organizations. - Gestalt of the grid, Performance Evalua- tion and Characterization of Parallel and Distributed Computing Tools. - (2002) pyGlobus – A CoG Kit for Python, http://www-itg.lbl.gov/gtg/projects/pyGlobus/.. - http://www.globus.org/cog/documentation/papers/.. - Proceedings of the 6th IEEE Symposium on High-Performance Distributed Computing, August pp. - Myproxy, Proceedings of the Tenth International Symposium on High Performance Distributed Computing (HPDC-10). - Proceedings of the 11th Symposium on High Perfor- mance Distributed Computing, Scotland, Edinburgh, July, 2002, pp. - http://www.globus.org/cog.. - The Physiology of the Grid: An Open Grid Services Architecture for Distributed Systems Integration, http://www.globus.org/research/papers/ogsa.pdf.
Xem thử không khả dụng, vui lòng xem tại trang nguồn hoặc xem
Tóm tắt