Grid Middleware. Gregor von Laszewski and Kaizar Amin. Chapter 5: Middleware for Communications, ed. Qusay H. Mahmoud. Wiley Publishers 2004. (pdf)

Abstract:

The creation of Grid middleware is an essential aspect of the software engineering effort to create an evolving layer of infrastructure residing between the network and applications. In order to be effective, this Grid middleware manages security, access, and information exchange to

• develop collaborative approaches for a large number of users, including developers, users, and administrators, that can be extended to the larger set of Internet and network users;

• transparently share a wide variety of distributed resources, such as computers, data, networks, and instruments, among the scientific, academic, and business communities; and

• develop advanced tools to enable, manage, and expedite collaboration and communication.

Scenarios that influence the state-of-the art middleware development include challenging worldwide efforts to deliver scientific experiment support for thousands of physicists at hundreds of laboratories and universities. Such an advanced environment is possible only if the middleware supports high-end and high-performance resource requirements to allow managing and analyzing petabytes of data while using end-to-end services under quality and security provisions.

In this chapter we present information about research activities to develop such computing environments that are supported through Grid middleware. We address the following questions:

• What does the term Grid mean?

• How does the Grid middleware help to develop a collaborative science and business environment?

• What trends exist in Grid middleware development?

• What applications exist that use Grid middleware?

• Where can we find more information about Grids?

• What will the future bring?

Quality of Service Based Grid Communities. Omer Rana, Asif Akram, Rashid Al-Ali, David Walker, Gregor von Laszewski, and Kaizar Amin. Chapter 8: Extending Web Services Technologies, The Use of Multi-Agent Approaches. Ed: Lawrence Cavedon, Zakaria Maamar, David Martin and Boualem Benatallah, 2004, pages 161-186.

Abstract:

Locating suitable services within a dynamic distributed system is a computationally intensive process, with no guarantee of quality and suitability of the discovered services. This is especially true for transient services, i.e. services which are likely to exist over short time frames. A significant effort has already been spent on developing distributed registry systems – such as the UDDI registry in Web services – to enable services to be published, and subsequently discovered. Regardless of the type of registry being used, it is nevertheless important to categorise services based on their particular properties – a process that should also aid the subsequent discovery of the service. A mechanism for grouping services based on a particular set of properties is investigated here – leading to the formation of service communities. Each such community is based on the existence of common parameter values being shared by members of the community. The structure of such a community is described, and a particular type of community established on the basis of Quality of Service properties is subsequently developed.

A Model for Quality-of-Service Provision in Service Oriented Architectures. Rashid Al-Ali, Omer Rana, Gregor von Laszewski, Abdelhakim Hafid, Kaizar Amin and David Walker. Journal of Grid and Utility Computing, Inderscience (2005).  (part1/part2)

Abstract:

Clients of service-oriented Grids are often concerned with service “quality” and the computational/economic costs for using such a service. A Quality-of-Service (QoS) mechanism should identify the resources needed to execute a service at a specified quality level.  It is also important that the selection of such resources is undertaken subject to other constraints, such as increasing/decreasing the costs associated with service execution on a particular set of resources. The QoS models presented here is based on distinguishing between a service provider, a service and a resource. Resources are entities that can be reserved, and a Service-Level-Agreement (SLA) presents a contract that needs to be agreed between a client and a service provider prior to resource allocation. The problem is to determine, given multiple types of QoS requests from clients, the optimal resource allocation to maximize utilization and maintain requested quality levels.  A model for such QoS provision is presented, along with a description of optimization heuristics. A QoS system which implements the model, and also involves support for resource reservations in advance is proposed. The implementation extends the functionality of the CoG Core toolkit.

Network QoS Provision for Distributed Grid Applications. Rashid Al-Ali, Shaleeza Sohail, Omer Rana, Abdelhakim Hafid, Gregor von Laszewski, Kaizar Amin, Sanjay Jha and David Walker. International Journal of Simulations,  Vol. 5 No 5. ISSN 1473-804x online, 1473-8031 print. 13. (pdf)

Abstract:

Grid computing provides a global-scale distributed computing infrastructure for executing scientific and business applications. An important requirement is the need to make this infrastructure appear as a single logical coordinated resource – which in reality consists of a variety of resources aggregated across different administrative domains. The aggregation of network resources is often undertaken over a ‘best effort’ infrastructure as provided by the Internet – however many applications, which necessitate soft-real time constraints, such as collaborative working or remote visualisation, require more stringent traffic guarantees. Managing Quality of Service (QoS) requirements across these aggregated resources therefore becomes an important concern. Such QoS criteria must extend to computational, data and network resources, and are often expressed in a Service Level Agreement, multiples of which may co-exist over the entire collection of resources. In this paper we focus on network QoS as part of our wider work on the Grid QoS Management (G-QoSm) framework. Provisioning of such QoS support is provided via the Differentiated Services architecture, and relies on the use of a Bandwidth Broker (BB) component. Performance results of using the BB alongside other elements in G-QoSm are presented. 

Ad hoc Grid Security Infrastructure. Amin, K.; von Laszewski, G.; Sosonkin, M.; Mikler, A.R.; Hategan, M.; Grid Computing, 2005. The 6th IEEE/ACM International Workshop on 13-14 Nov. 2005 Page(s):8 pp. 

Abstract:

This paper describes an ad hoc grid security infrastructure developed as a part of the Java CoG Kit project. It supports several requirements specific to the sporadic nature of ad hoc grids. It focuses on identity management, identity verification, and authorization control in spontaneous grid collaborations without pre-established policies or environments. It adopts established community standards, with modifications where needed. This paper also discusses the integration of the ad hoc grid security infrastructure in an ad hoc grid implementation. The implementation supports secure collaboration in ad hoc grids using commodity technologies such as the Java CoG Kit, JXTA, GSI, and XACML.

An Abstraction Model for a Grid Execution Framework. Kaizar Amin, Gregor von Laszewski, Rashid Al Ali, Omer Rana, and David Walker, Journal of Systems Architecture, Volume 52, Issue 2 , February 2006, Pages 73-87, Parallel, Distributed and Network-based Processing. (pdf)

Abstract:

Computational Grids have been identified as one of the paradigms revolutionizing the discipline of distributed computing. The contributions within the Grid community have resulted in new Grid technologies and continuous improvements to Grid standards and protocols. Though crucial to the success of the Grid approach, such an incremental evolution of Grid standards has become a primary cause of frustration for scientific and commercial application communities aspiring to adopt the Grid paradigm. Motivated by our rich experience and the need to decouple the application development and the Grid technology development processes, we propose an abstractionbased Grid middleware layer as part of the Java CoG Kit. In this paper, we showcase our abstraction model and verify its extensibility by integrating it with an advanced quality-of-service–based execution framework.

Quality Assured Ad Hoc Grids. Kaizar Amin, Gregor von Laszewski, and Armin R. Mikler. In International Conference on Autonomic and Autonomous Systems International Conference on Networking and Services. IEEE, 23-28 October 2005. (pdf)

Abstract:

This paper presents an integrated architecture for ad hoc Grids developed within the Java CoG Kit project. It provides an overview of the key component frameworks that collectively build the ad hoc Grid architecture. Further, it outlines a formal model that can be formally evaluated. The paper also presents an enhancement to the Java CoG Kit to address requirements posed by ad hoc Grids. It integrates into the Java CoG Kit commodity technologies such as Jxta, and ClassAds.

An Overview of Grid File Transfer Patterns and their Implementation in the Java CoG Kit. Gregor von Laszewski, Jarek Gawor, Pawel Plaszczak, Mike Hategan, Kaizar Amin, Ravi Madduri, and Scott Gose. Journal of Neural Parallel and Scientific Computing, 12(3):329-352, September 2004. Special Issue on Grid Computing. (pdf)

Abstract:

Accessing files on remote resources is a required function in Grids. In this paper, we report on the file transfer patterns supported in the Java CoG Kit. These patterns are supported by a rich set of accompanying components, including Java classes and methods, command line tools, graphical user interfaces, and portals. The patterns and their implementations are exposed through familiar Java language capabilities of interfaces, hence hiding the underlying protocols. Using these interfaces, one can provide a variety of implementations for diverse file transfer mechanisms and protocols. Together, these tools can be used to implement more sophisticated services. We present a number of prototype applications reusing the Java CoG Kits file transfer patterns. Additionally, we present performance numbers based on a typical client deployment scenario. 

Analysis and Provision of QoS for Distributed Grid Applications. Rashid AlAli, Kaizar Amin, Gregor von Laszewski, Omer Rana, David Walker, Mihael Hategan, and Nester Zaluzec. Journal of Grid Computing, 2(2):163-182, June 2004. (pdf)

Abstract:

Grid computing provides the infrastructure necessary to access and use distributed resources as part of virtual organizations. When used in this way, Grid computing makes it possible for users to participate in collaborative and distributed applications such as tele-immersion, visualization, and computational simulation. Some of these applications operate in a collaborative mode, requiring data to be stored and delivered in a timely manner. This class of applications must adhere to stringent real-time constraints and Quality-of-Service (QoS) requirements. A QoS management approach is therefore required to orchestrate and guarantee the timely interaction between such applications and services. We discuss the design and a prototype implementation of a QoS system, and demonstrate how we enable Grid applications to become QoS compliant. We validate this approach through a case study of an image processing task derived from a nanoscale structures application.

Toward an Architecture for Ad Hoc Grids. Kaizar Amin, Gregor von Laszewski, and Armin R. Mikler. In 12th International Conference on Advanced Computing and Communications (ADCOM 2004), Ahmedabad Gujarat, India, 15-18 December 2004. (pdf)

Abstract:

The advantages offered by existing Grid frameworks have resulted in a wide range of applications adopting the Grid approach. The first generation of production Grids have focused on the creation of large virtual organizations that share high end resources as part of a static resource pool. However as many collaborative interactions take places on a sporadic or ad hoc fashion outside of the virtual organization, such Grids become impractical. In this paper, we outline an extension to the Grid architecture that addresses this issue. We refer to this architecture a as sporadic or ad hoc Grid. We discuss use cases that justify our efforts toward a self-organizing ad hoc Grid architecture. We outline the functional principles of this architecture and propose our framework to implement them.