1,720,963 research outputs found
How to run scientific applications over web services
No full textToday, the task of running and coordinating a scientific application across several administrative domains is extremely complex. As an example, the most popular tool for scientific applications, MPI, is not designed to address firewall limitations or data heterogeneity, even if its extensions deal with some of these problems. In this paper, we design a new approach to run a scientific application in a distributed environment, when data and computing power are scattered across the Web: Web Services can be used to tunnel computation and data migration. We show that a very simple mapping exists between MPI primitives and the Web Service infrastructure. We are currently designing a framework, based on Web Services, which will implement the main MPI primitives: this way an MPI application could be run on any platform supporting Web Services. © 2005 IEEE
The social network of Java classes
No full textSeveral works in literature have analyzed the link structure of programs in relation with software engineering: it has been observed that the programming standards caused small-world networks to emerge among classes in object-oriented programming. The need for coherent design and the coding conventions introduce regular patterns in the link structure of code. In this work, we study the social network naturally emerging from unrelated software projects. We studied the links present among Java classes coming from different contexts. In this case, any observable patterns come from social behaviors, rather than software engineering practices. In our analysis, we could observe a regular social network, organized according to a power-law distribution that is typical, for instance, of links among Web pages. We give a positive value to class links, which we consider a sign of relevance and acceptance. Out of this, we propose a way of ranking classes, and we present our prototype search engine for Java classes. Copyright 2006 ACM
A search architecture for grid software components
No full textThe use of a search architecture for grid software components that allowed the design of workflow-based Grid applications was discussed. The main issue presented was the component search service that allowed users to locate the components which they needed. The search engines, that were based on information retrieval techniques, were able to rank components on the basis of their similarity with the place-holder description, their popularity among developers and their use within other services. The use of such tool can help the whole Grid to become a virtual machine, tapping the power of a vast number of resources
Load-balancing and caching for collection selection architectures
No full textTo address the rapid growth of the Internet, moder Web search engines have to adopt distributed organization where the collection of indexed documents is partitioned amon several servers, and query answering is performed as a parall and distributed task. Collection selection can be a way to reduc the overall computing load, by finding a trade-off between th quality of results retrieved and the cost of solving queries. I this paper, we analyze the relationship between the collectio selection strategy, the effect on load balancing and on the cachin subsystem, by exploring the design-space of a distributed searc engine based on collection selection. In particular, we propose strategy to perform collection selection in a load-driven wa and a novel caching policy able to incrementally refine th effectiveness of the results returned for each subsequent cach hit. The combination of load-driven collection selection an incremental caching strategies allows our system to retrieve tw thirds of the top-ranked results returned by a baseline centralize index, with only one fifth of the computing workload
Toward GRIDLE: A Way to Build Grid Applications Searching Through an Ecosystem of Components
No full textUsing Web Services to Run Distributed Numerical Applications
No full textMPI is a de facto standard for high performance numerical applications on parallel machines it is available, in a variety of implementations, for a range of architectures, ranging from supercomputers to clusters of workstations. Nonetheless, with the growing demand for distributed, heterogeneous and Grid computing, developers are hitting some of its limitations e.g. security is not addressed, and geographically distributed machines are difficult to connect. In this work, we give an example of a parallel application, implemented with the use of Web Services. Web Services represent an emerging standard to offer computational services over the Internet. While this solution does not reach the same performance of MPI, it offers a series of advantages high availability, rapid design, extreme heterogeneity. © Springer-Verlag 2004
Component metadata management and publication for the grid
Get PDFThere is growing attention to component-oriented software design of Grid applications. Within this framework, applications are built by assembling together independently developed software components. Two main approaches are commonly used to manage, develop and publish software components: one is based on an Interface Description Language (IDL); the other is typical, for instance, of Java and is based on introspection and design conventions. In this paper, we compare them and we propose a third approach that merges the flexibility and fast learning curve of the latter, with the rigor of the former. Our proposal is meant to help the transition towards more modern tools, which is required to develop versatile Grid applications. © 2005 IEEE
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
