Browsing by Author "Xia, Jizhe"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item A New Model for a Carpool Matching Service(Public Library of Science, 2015-06-30) Xia, Jizhe; Curtin, Kevin M.; Li, Weihong; Zhao, YonglongCarpooling is an effective means of reducing traffic. A carpool team shares a vehicle for their commute, which reduces the number of vehicles on the road during rush hour periods. Carpooling is officially sanctioned by most governments, and is supported by the construction of high-occupancy vehicle lanes. A number of carpooling services have been designed in order to match commuters into carpool teams, but it known that the determination of optimal carpool teams is a combinatorially complex problem, and therefore technological solutions are difficult to achieve. In this paper, a model for carpool matching services is proposed, and both optimal and heuristic approaches are tested to find solutions for that model. The results show that different solution approaches are preferred over different ranges of problem instances. Most importantly, it is demonstrated that a new formulation and associated solution procedures can permit the determination of optimal carpool teams and routes. An instantiation of the model is presented (using the street network of Guangzhou city, China) to demonstrate how carpool teams can be determined.Item Optimizing Access to Big Earth Observation Data with Spatiotemporal Patterns -- An Example with the GEOSS Clearinghouse(2015) Xia, Jizhe; Xia, Jizhe; Yang, ChaoweiBig Data becomes increasingly important in almost all scientific domains, especially in geographical studies where millions to billions of sensors are collecting data of the Earth continuously. Recognizing the importance of managing the Big Earth observation Data, Group on Earth Observations selected the Global Earth Observation System of Systems Clearinghouse (CLH) to harvest, manage and share Earth observation metadata. Building a CLH to support global operation is very challenging, because it is essential for CLH to effectively manage and index Big Earth observation Data, provide accurate data service evaluation, and execute these services using fast provision computing resources to different space and time locations to support dynamic global user access. Although various optimization mechanisms (e.g., index, workload balancing, service model, cache) have been proposed, few approaches optimize the Earth observation data access with the spatiotemporal patterns of the data utilization. This dissertation investigates a variety of spatiotemporal optimizations to better support Big Earth observation Data access using the CLH as an example. Specifically, the objectives are the following: (1) develop a new indexing mechanism to accelerate Big Data access. The new indexing mechanism integrates the spatiotemporal user access patterns into traditional index structures. The experiment result showed that the new index yields 9-20% performance gain for the data access compared to a classic R*-tree index; (2) develop a new service performance model to improve the service evaluation accuracy. The new service model collects globally distributed service information with cloud services and volunteers, and integrates the spatiotemporal service characteristics to provide evaluation end users at different space-time locations. The proposed spatiotemporal service model yields 3-18% accuracy improvements gains, thereby helping end users better choose service for data access; and (3) develop a cloud computing adoption framework to better support global user access and spiking access. The cloud framework automatically provisions and delivers computing resources for different data access tasks with spatiotemporal computing workloads, and globally deploys system instances to different regions. The experiment result showed that the cloud framework helps the CLH achieve about 10 seconds’ performance gains for global and spiking user access. The significance of this research is that it provides a potential solution for optimizing access to Big Earth observation data using spatiotemporal data utilization patterns, thereby better supporting various Big Data related studies with faster data access.Item Optimizing an Index with Spatiotemporal Patterns to Support GEOSS Clearinghouse(2013-02-18) Xia, Jizhe; Xia, Jizhe; Yang, ChaoweiBig Data becomes increasingly important in almost many scientific domains, especially in geographic science where hundreds to millions of sensors are collecting data of the Earth continuously (Whitehouse News 2012). The data are managed and served through various Geospatial Cyberinfrastructure (GCI) components worldwide, and many GCI components are also developed to help discover and utilize the widely geographically dispersed data. In the Internet Era, users expect to receive responses in seconds for the discovery and it is a big challenge to achieve it with a proper index. For example, the R-tree (Guttman 1984) leverages spatial relationship among features and is widely used in spatial DataBase Management Systems (DBMSs) and different R-tree variants have been proposed to 1) improve data retrieval performance, 2) support temporal indexing, and 3) utilize multiple computers for indexing. However, it is hard to meet the seconds expectation because little research has included spatiotemporal patterns of user queries. Traditionally, user behavior has rarely been considered in a spatial index and only one single index is used to support all users from different regions at different times. I propose a Predefined Multiple Indices Mechanism (PMIM) to support global user queries by predefining different indices for different categories of users who have similar query patterns. Access Possibility R-tree (APR-tree) is proposed to build an index based on spatiotemporal patterns of user queries. The new spatiotemporal indexing strategy provides a potential solution to leverage Big spatial Data indexing and enable seconds response to global users. Using metadata in the GEOSS Clearinghouse as an example, I conducted a series of performance experiments for PMIM implemented using APR-tree. Experiment results indicate that new indexing mechanism outperforms a regular R*-tree.