Social media is often useful for discovering contextual information that is difficult to find on traditional query-based search engines such as Google. For example, temporal events such as traffic incidents are often posted on social media due to the wide-reaching and real-time nature of social media platforms. Social media can also be used to model the sentiments and opinions of different geographical regions, as well as provide a platform for organizing social movements. In this paper, we give an overview of our system, GeoContext, which models a stream from Twitter into topics and analyzes the geographical locations of the topics. GeoContext includes methods for filtering a social media stream by keywords and location coordinates in order to provide more specific topics. GeoContext includes a geolocation module, called GeoContext Locator, for predicting the locations of tweets that are not associated with explicit coordinates, in order to model topics in different locations.

BACKGROUND: Rapid improvements in light and electron microscopy imaging techniques and the development of 3D anatomical atlases necessitate new approaches for the visualization and analysis of image data. Pixel-based representations of raw light microscopy data suffer from limitations in the number of channels that can be visualized simultaneously. Complex electron microscopic reconstructions from large tissue volumes are also challenging to visualize and analyze. RESULTS: Here we exploit the advanced visualization capabilities and flexibility of the open-source platform Blender to visualize and analyze anatomical atlases. We use light-microscopy-based gene expression atlases and electron microscopy connectome volume data from larval stages of the marine annelid Platynereis dumerilii. We build object-based larval gene expression atlases in Blender and develop tools for annotation and coexpression analysis. We also represent and analyze connectome data including neuronal reconstructions and underlying synaptic connectivity. CONCLUSIONS: We demonstrate the power and flexibility of Blender for visualizing and exploring complex anatomical atlases. The resources we have developed for Platynereis will facilitate data sharing and the standardization of anatomical atlases for this species. The flexibility of Blender, particularly its embedded Python application programming interface, means that our methods can be easily extended to other organisms.

The long term storage of data has been a continuing concern within the supercomputer community. In fact, the supercomputer community has continually pushed the forefront of mass storage systems. Mass storage currently implies the storage of data in excess of 1 0 12 bits and it is not uncommon in this community to see requirements of over 1015 bits. We have developed a methodology for describing user requirements in a common format so that a number of workloads for a significant portion of the Intelligence Community can be compared easily. The common format is a list of questions, which includes data that is useful in modeling at a very high level. TIvo very diff erent but rep­ resentative workloads are presented. We have also identi­ fied areas of research and development in mass storage systems that are important to the Intelligence Community.

The week-end production period on a Cray X-MP was monitored for several months at each of two supercomputing sites. The hardware performance monitor available on the X-MP was used to collect the data at each site. Various metrics are computed using the measured data. These metrics include rates, such as mops mflops, percent vectorization, average vector lengths, memory reference rates, memory contention, the operation mix, I/O rates, hardware metrics such as operations per clock period, and percentage of time waiting to issue an instruction that is waiting on some hardware component. These metrics are used in evaluating the performance of the workloads and the X-MP architecture. These metrics may also give insight into the design of similar register-based architectures and may be useful in parameterizing models in future work.>