Status Report for 11/29/2006

In the light of our last discussions and your comments,
here is the new 2-page of *draft* abstract

http://complexity.ucs.indiana.edu/~asayar/proposal/newAbstract1.doc

I am going to put the capability based data+service integration issues a side and I will keep this problem in my mind as my long term research topic.

My current research focus is based on what I have done so far which is summarized in my previous post.

In summary, my research consists of the distributed map processing and rendering of the scientific data and information from the interoperability, scalability and performance points of view

Thanks for your comments in advance.

Summary of what I have been doing

Here in this snapshot you see the summary of my work done so far.

I am going to upload the complete presentation when I am done.

I might also enhance the outline with the definitions, motivations, research issues, and the "title" which is the most important one.

Introducing Capability Concept into Service+Data Integration

Here are the expected presentation slides. I gave the outline in my previous post.

In these slides (link is given below), I summary my project about introducing capability concept into the data+service integration problems.

Capability concept is first introduced by OGC (Open GIS Consortium) in GIS (Geographic Information Systems) domain. Their definition of capability is specific to geographic data (which is called feature) and provides ad-hoc solutions to "data access" and "service-chaining" issues in GIS domain.

The term "capability" is also used in Grid community as a tool enabling fine-grained authorization for general Web/Grid services in compliant with Web Services Specs.

http://complexity.ucs.indiana.edu/~asayar/proposal/capabilityConcept2.ppt

Status Report for 11/16/2006

I have been preparing slides titled "Introducing Capability Concept for the Data+Service Integration".

For the concept of the slides please see the "outline" listed below.
I have finished almost everthing except for the most challenging part. This part is titled as "Research Issues". I will upload the slides as soon as I am done with this part.

Please let me know your comments and questions.

Comparing ASIS with the current data integration approaches

Here is the slides about comparing our data-integration approach with 3 other well-known approaches. These are SRB (Storage Resource Broker), Ogsa-DAI (Ogsa-DAI) and Sompel's Digital Libraries. We call our approach ASIS (Application Specific Information Systems) and it is based on OGC's capability-based service chaining.

http://complexity.ucs.indiana.edu/~asayar/proposal/dataIntegrationApproaches2.ppt

Distributed scientific and engineering applications require support for integrating data from multiple data sources through distributed query and transactions facilities. Data resides in multiple forms in numerous heterogeneous repositories and served by various kinds of data servers. In my presentation I mentioned about the data integration approaches and comparing our capability-based Application-Integration Framework (ASIS) with the other approaches.

IVOA's Approach to Data Integration

Please see the slides presenting IVOA's approach to "Data Integration" issues. http://complexity.ucs.indiana.edu/~asayar/proposal/IVOA_DataIntegIssues.ppt

The goal of the IVOA (International Virtual Observatory Alliance) is the development of architectural decisions and standards in the astronomy domain

IVOA is not directly addressing the data integration issues, but they plan to define VOStore and VOSpace based on SRB and/or NGAS architectures. IVOA has a prototype application for the data integration, OpenSkyQuery. OpenSkyQuery is VO prototype application that marries Web Services technology with emerging VO standards to enable dynamic cross-matching queries between different VO-enabled archives. In my presentation, I tried to explain basically OpenSkyQuery and related IVOA standards and specifications.

Status Report for 10/31/2006

My current and future work outlined in one ppt slide :
http://complexity.ucs.indiana.edu/~asayar/proposal/capabilityConcept.ppt

Our implementation of data integration is based on geographic data model (GML) and online services (WMS and WFS) whose standard specifications are defined by OGC.

The OGC specifications are mostly domain specific and fine-grained. The specifications are dynamic and keep changing based on the communities needs and new technologies. Therefore, even in the same domain and same science community, it is very hard to create a generic capability. In order to get a rough idea about the OGC-WMS capability and how deeply it is GIS specific, please see this link (http://schemas.opengis.net/wms/1.3.0)

For example, in order to be able to apply OGC like capability based data+service integration architecture to Chemistry, a new capability schema specific to Chemistry data model (CML) should be defined by a standard body in the Chemistry community. If the community does not have the capability definitions (which is normal), and If we want OGC like capability based data+Service integration we need to define a generic (or in Chemistry domain) capability.

In order to get rid of this burden of creating different capabilities and defining different services for different domains we try to create a generic capability and architectural framework to integrate data+service in general science domains. So, these issues might be our future research.