Hi All,
Midterm is approaching! Well, most of the students won’t be happy to hear that, but anyway, we achieved couple of things in the first half and we’ll try to accomplish more in the second half.
This week, we took a look at the First Provenance Challenge workflow trace and example traces from VisTrail workflow system both in .xml format. Victor has wrote a xslt program that can transform .xml file to a .csv file and he also helped me figure out the issues with the RPQ engine, thanks for that. Upon understanding the meaning of trace files in .xml format, we’re now able to extract useful information for the file and transform that into a .csv file, which is capable for our system. However, some of the queries can’t be (at least not easy) implemented in RPQ, so I’l like to propose a question: should we develop another system to answer those queries? What about the overhead and how to categorize different queries and feed them into different systems? We’ll discuss that next week.
Couple of things left to do: 1. Since every user of our system need to setup the Postgres database locally, which is not a trivial task in some case, so we could to establish the database on the server and free up the user from setting that up. 2. Yaxing has updated a document about the EVA use case, we’ll take a look at it and try to get some examples out of it. 3. Next Friday, I’ll present the paper of RPQ on large graph to the lab group. 4. The Google site of ProvWG is under construction, hopefully by the end of the summer program, it’ll show the outcome of this summer.
Best,
Tianhong