University of Washington

The team at the University of Washington is working with the University of Utah on building a new infrastructure for computational oceanography that uses the Google/IBM cloud to allow ad hoc, longitudinal query and visualization of massive ocean simulation results at interactive speeds. In addition their Astronomy Survey Group is conducting research to Scale the Sky with MapReduce/Hadoop.

Research Projects
Resources

Research Projects

Scaling the Sky with MapReduce/Hadoop

Astrophysics is addressing many fundamental questions about the nature of the universe through a series of ambitous wide-field optical and infrared imaging surveys. New methodologies for analyzing and understanding petascale data sets are required to answer these questions. This research project is focused on developing new algorithms for indexing, accessing and analyzing astronomical images. This work is expected to have a broad range of applications to other data intensive fields.

Where the Ocean Meets the Cloud

This project is building a new infrastructure for computational oceanography that uses the CluE platform to allow ad hoc, longitudinal query and visualization of massive ocean simulation results at interactive speeds. This infrastructure leverages and extends two existing systems: GridFields, a library for general and efficient manipulation of simulation results and VisTrails, a comprehensive platform for scientific workflow, collaboration, virtualization, and provenance.

Resources

Presentation: Towards Interactive Visualization in the Cloud
October 22 2009
Towards Interactive Visualization in the Cloud

Video: An Interview with Bill Howe of the University of Washington
August 24 2009
Visualizations have played an important role in science for decades. As data collection becomes more sophisticated, finding ways to visualize large data sets in a relatively easy and affordable manner has become an imperative. As part of the Clue Initiative, Bill Howe, a computer scientist at the University of Washington, is working with colleagues at the University of Utah to develop a cost effective visualization tool based on the VisTrails suite.

Paper: CloudViews: Communal Data Sharing in Public Clouds
July 22 2009
By Roxana Geambasu, Steven D Gribble, and Henry M Levy. Abstract: This paper focuses on a new opportunity introduced by the cloud environment: specifically, rich data sharing among independent Web services that are co-located within the same cloud. In the future, we expect that a small number of giant-scale shared clouds – such as Amazon AWS, Google AppEngine, or Microsoft Azure – will result in an unprecedented environment where thousands of independent and mutually distrustful Web services share the same runtime environment, storage system, and cloud infrastructure. One could even imagine that most of the Web will someday be served from a handful of giant-scale clouds. In particular, we argue that co-location creates an auspicious environment for Web service composition, which in turn spawns immense opportunities for simplifying Web service construction. Three key technological features differentiate the shared-cloud world from the traditional in-house datacenter and enable these opportunities: (1) free, efficient, and plentiful network bandwidth that supports tighter and larger-scale Web service integration than is possible over wide-area networks; (2) a shared storage system that can provide powerful abstractions for convenient, efficient, and large-scale inter-service data sharing; and (3) the potential for a rich runtime ecosystem consisting of many “utility” Web services that act as building blocks for other services and facilitate their implementation greatly. This paper is divided into two parts that describe the opportunities for collaborative Web services in the new cloud world. It also examines the three technological features described above and both the potentials and the challenges that they bring. Section 3 presents CloudViews, a cloud storage system we are developing to facilitate collaboration through protected inter-service data sharing. CloudViews is one example of the kinds of functions public clouds must offer to facilitate Web service development. Overall, our goal is to provide a high-level glimpse of the potential impact of co-location on low-level cloud infrastructure and service development.