GLIDE: A Grid-Based Light-Weight Infrastructure for Data-Intensive Environments

Abstract

Abstract. The promise of the grid is that it will enable public access and sharing of immense amounts of computational and data resources among dynamic coalitions of individuals and institutions. However, the current grid solutions make several limiting assumptions that curtail their widespread adoption in the emerging decentralized, resource constrained, embedded, autonomic, and mobile (DREAM) environments: they are designed primarily for highly complex scientific problems, and therefore require powerful hardware and reliable network connectivity; additionally, they provide no application design support to grid users (e.g., scientists). To address these limitations, we present GLIDE, a prototype light-weight, data-intensive middleware infrastructure that enables access to the robust data and computational power of the grid on DREAM platforms. We illustrate GLIDE on an example file sharing application. We discuss our early experience with GLIDE and present a set of open research questions.

A software architecture-based framework for highly distributed and data intensive scientific applications

“…It also allows us to introduce flexibility in the individual instances of the reference architecture while, at the same time, controlling the legal system configurations. Finally, the explicit connectors and well-defined component interfaces allow OODT in principle to integrate with a wide variety of third-party systems (e.g., [26]). The outcome of the guiding principles (described above) and design decisions (detailed below) is an architecture that is "easy to build, hard to break".…”

Deployment of Infrastructure and Services in the Open Grid Services Architecture (OGSA)

“…Problems with deployment in Globus have been documented [26], as have Grid deployment Use Cases which complement our deployment scenarios [27,28]. Other approaches and tools for Grid deployment include GITS [29], distributed Ant [30], the IBM autonomic deployment framework [31], deployment planning [32], PACMAN [33], GPT [34], and Virtual Machines [35].…”

The GridLite DREAM: Bringing the Grid to Your Pocket

“…GridLite is an adaptable, light-weight grid platform that is equipped with the appropriate support for application design, implementation, deployment, and evolution. To realize GridLite, we have drawn upon our experience in the areas of distributed computing [17], software architecture and in particular its role in DREAM environments [21], and distributed data management [19]. In this section, we describe the architecture of GridLite, including the three key objectives that motivated its separation into three distinct layers.…”

“…The services include discovering resources on the grid (such as data or computation provision), monitoring grid and grid node performance, wrapping structured resources to enable access in a heterogeneous setting, dynamically "morphing" deployed system architectures, and deploying uniform grid software components in the face of different execution platforms. Experience in resource and metadata description [19] and software architectures [17] is necessary to enable GLS to utilize GLM effectively and efficiently.…”

“…Our position is that, in order to address the aforementioned limitations of the grid paradigm, a new type of software solution must be constructed. To that end we propose, and have implemented an early prototype of, GridLite, a software architecture-based grid platform suitable for deployment in DREAM environments [19]. GridLite is heavily influenced by, and its early design and implementation directly rely on, our OODT data grid [18] and Prism-MW architectural middleware [17] platforms.…”