DMQ4: Online remote access to instruments

Visualization Tools

The instrument data visualization tools are being developed as part of the DART DMQ4 workpackage. The objective of DMQ4 is to provide online, remote access to pilot working instruments and sensors. This objective involves:

• Developing visualization tools for data collected from instruments and sensors - to provide the user with a friendly interface to request past data in a particular period or satisfying some constraints.
• Data processing and data mining for living data streams collected from instruments and sensors - to monitor and analyze especially interesting or emergent data from extensive data streams on-the-fly.

To implement the visualization tools, serveral advanced techniques and softwares are involved:

• Common Instrument Middleware Architecture (CIMA) -- CIMA, developed by Indiana University, is a framework for making instruments and sensors network accesssible in Internet. DMQ4 utilizes CIMA to import instrument/sensor data.
• Storage Resource Broker (SRB) -- SRB, developed by the San Diego Supercomputing Center, provides a uniform interface to hetergenous data storage resources over a network. DMQ4 adopts SRB to store and manage instrument/sensor data.
• Synchronized Multimedia Integration Language (SMIL), Scalable Vector Graphics (SVG) and HTML+TIME (Timed Interactive Multimedia Extensions for HTML). DMQ4 integrated these W3C technologies for choreographing multimedia presentations in the web browser for users.

Technologies

The new DMQ4 tools are supported on Internet Explorer version 6 and above and Firefox 1.5.x with ActiveX.

• Languages: Java, Javascript, SPARQL
• Java libraries: Sesame, Elmo, Java Server Faces (JSF)
• Javscript libraries: Dojo, MochiKit, Plotkit
• Servers: SRB, Fedora, Tomcat
• Web technologies: Dynamic HTML, SVG/HTML Canvas (in supporting browsers), XHTML, ActiveX
• Media Player: Windows Media Player

Architecture

1. The user initiates a search query for a given experiment, project and date range.
2. The DMQ4 web application generates SPARQL queries and performs the query against the RDF taken from the Fedora server.
3. The DMQ4 web application processes the results RDF and uses the Elmo library to create Java objects.
4. The resultant Java objects are processed using JSF to generate the required Javascript and HTML to be rendered by the client's web browser.
5. The client processes the results of their query. The results contain URLs to the Fedora resources. These requests are processed by Fedora which accesses the underlying SRB server to deliver the sensor data and image files.

Previous Prototype

The previous prototype was based the following technologies:

• Programming Languages: Java, JSP, Java Script.
• Data Storage: SRB (License: BSD)
• Data Importation: CIMA
• Web Server: Tomcat (License: Apache)
• Web Design: DHTML, SVG, HTML+TIME

Screen Captures

Laboratory selection

Input validation

Instrument selection

Time selection

Query result

No data

Software

The DMQ4 Online instruments/sensors visualization tool is made up of the following components:

• An interactive rendering of query results
• Servlet based user interface to submit queries and display the results
• A query layer to allow users to submit location and time ranges to return sensor results

To install the software:

• Download and install Tomcat 6.0.4.
• Download the DMQ4 Online Instruments/Sensors Visualization Tools.
• Unpack the zip file and copy the following files into the lib directory of Tomcat:
  • el-ri.jar
  • jsf-1.2-api.jar
  • jsf-1.2-impl.jar
  • jstl-1.2.jar
• Edit the tomcat-user.xml and add:
  <role rolename="manager"/>
  <user username="admin" password="" roles="manager"/>
• Run the webapp-deploy target from ant to deploy the WAR file. It is expecting the default configuration options (localhost:8080/manager) for deployment. See the build-tomcat.xml to change any of these values.