airavata-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pamidighantam, Sudhakar V" <>
Subject Re: DataCat Project Progress
Date Thu, 11 Dec 2014 14:12:41 GMT
I support these goals. I am available for you to engage with you and anything I can do to
expedite the project please let me know. Even if you think I can not,  do please ask anyway.
I have written the original parsers in perl myself and have directed others when the Cup/JFLex
system was put in place. I have generated and modified the CUP/JFlex code before, so I am
familiar with how it works. I will look at the paper and may suggest  additions.

I can not test the system now by adding more data and see if this can parse the new data.
How can we get to that point. This is critical first step for me before I can ask friendly
users to test this further. Current state is a prototype only and is not interesting enough
for any of our users. Unless we can add parsing of more data and more salient data and create
products it is difficult engage end users in any meaningful way.

Perhaps if this is deployed somewhere in Indiana it may be easier to move forward. If you
need more data please let me know where I should locate it for you to access.


On Dec 11, 2014, at 4:11 AM, Supun Nakandala <<>>

Hi All,

We had the mid evaluation of the project last Tuesday and the following concerns were raised.

  1.  The lack of visibility of the overall solution in the project demonstration.
  2.  The ability to come up with a solution where, scientist who does not have a background
in computer science can create new parsers (metadata extraction logic)

The project was demonstrated using the web interface that we developed. For the final evaluation
we expect to demonstrate the system using laravel PHP Reference Gateway running in a production
server and demonstrate how a new data product that gets generated will be identified, indexed
and will be available for searching and hope this will handle the first issue.

We also had a meeting with Dr. Dilum our internal supervisor where we identified things that
can be done from now to 15th January, the expected project completion date

  1.  Do a proper performance test and publish a paper before final marks for the project
is finalized (marks will be finalized by the end of March).
  2.  Getting to work more parsers, so that Sudhakar can ask more users to use the system.
This will help to get more feedback on the system and have a real world usage.
  3.  Implement the support for provenance aware workflow execution in Airavata using our

We have written a draft paper which I have attached here with. We showed this to Dr. Srinath
and Dr Dilum and they suggested that we do a proper performance testing (The one that already
done is not up to the expected standards). Given the available time we need to prioritize
our work and select a set of tasks that is doable and has the most impact. What do you all

Draft Paper:

Literature Review:


View raw message