Abstract Range query has a crucial role in large-scale data analysis. Unfortunately, the performance may be severely degraded by data skew. Such problem is often faced in large-scale parallel databases, peer-to-peer (P2P) systems as well as in Cloud computing. State-of-the-art methods designed to handle this problem offer significant improvements over naive implementations. However, performance could be further improved by reducing the cost of global... Range query has a crucial role in large-scale data analysis. Unfortunately, the performance may be severely degraded by data skew. Such problem is often faced in large-scale parallel databases, peer-to-peer (P2P) systems as well as in Cloud computing. State-of-the-art methods designed to handle this problem offer significant improvements over naive implementations. However, performance could be further improved by reducing the cost of global skew knowledge and broadcasting. Ganesan, Bawa and Garcia-Molina proposed a load-balancing algorithm that guarantees a good ratio between the maximum and minimum loads among nodes. However, their algorithm requires global max-min load information to use local load balancing operations. Global load information can be found with O (log n) messages. In order to reduce this cost, we propose OPTIMA, a novel online load balancing approach for range partitioned data. Whenever a partition becomes overloaded, data transfers are performed, in background, from the most loaded nodes to the least loaded ones as in Ganesan et al., work. As a result, the partition boundaries and data sizes change. The key point of our proposal is the imperfect knowledge of the global load information (partition statistics). We introduce the concept of “Imperfect Partitioning Vector” (IPV), where, both nodes and clients have an approximate information about the load distribution. They can nevertheless locate any data with almost the same efficiency as using exact partition statistics. Furthermore, maintaining load distribution statistics do not require exchanging additional messages or maintaining a data structure as opposed to the cost of efficient solutions from the state-of-art (which require at least O (log n) messages). Read More Read Less

Abstract The university research is known to the society in many forms out of which the news summaries and reporting play a major role. Scientific information reaches society at different levels and forms. This work reports and identifies how the university research output is published in two Japanese newspapers. We probed various issues such as comprehensive metrics, scores, referencing, domains and research cooperation during... The university research is known to the society in many forms out of which the news summaries and reporting play a major role. Scientific information reaches society at different levels and forms. This work reports and identifies how the university research output is published in two Japanese newspapers. We probed various issues such as comprehensive metrics, scores, referencing, domains and research cooperation during a specific period. Read More Read Less

Abstract The researchers increasingly interact with information resources normally and such interactions are characterized by the extensive search of the sources. The research also intended to address the issues and problems involved in information retrieval to provide an appropriate solution to their challenges. In this current exercise we have studied using a few qualitative framework. To supplement our work we have contacted them using... The researchers increasingly interact with information resources normally and such interactions are characterized by the extensive search of the sources. The research also intended to address the issues and problems involved in information retrieval to provide an appropriate solution to their challenges. In this current exercise we have studied using a few qualitative framework. To supplement our work we have contacted them using a structure set of questions to tap their access pattern. Our work reveals that the online resources become more indispensible in the few years. Also the social networks provide and support large number of users. More use of datasets and databases contribute to the networked information. The academic journals and the web pages of many libraries also contribute extensively. There are many issues and factors are studied by us in the investigation which includes title understanding and low correlation between titles and content. Researchers were to search for hours and days to find the information they needed. The Web environment, is still far from providing a well-organized information to information seekers and users. Lack of standardized format for searching information in search engines for databases lead researchers to misinformation, and most of interviewees complained about this matter. Results also showed that researchers categorize the retrieved information into four components: A. Explicit useful information, B. Hidden useful information, C. Explicit inappropriate information, and D. Explicit Disturbing information. Findings advise information systems policy makers to adjust their propositions on the information behavior of the new generation researchers and online residents and revise their indexing and collection development guidelines. Read More Read Less