Open sgosline opened 6 years ago
Talked to @meredithslota and @xschildw to get access to google analytics and opened a Jira ticket for warehouse access. It seems google analytics may be sufficient for basic data on views. Will confirm once I get access to it. Once I have access to the warehouse, @xschildw mentioned I'd be able to run queries directly on ~week old data (to avoid query-ing the live db).
Will take a look at @kdaily scripts as well (I assume to use/test them warehouse access is prerequisite too).
Correct! My scripts can query live or older data as well. I've worked with @kimyen extensively and the queries are not so intensive to affect data warehousing capabilities, but if more of us are running them we may need to revisit/use older data. I schedule my reports (currently using crontab
on an ec2, but have tested with AWS lambda) to run outside of @kimyen's existing import and processing time frames.
Going forward, using my ec2 is not preferred. We need to use the new scicomp-infra process (https://github.com/kdaily/scicomp-infra, led by @zaro0508) to get AWS resources for such things.
Got access to the warehouse DB just now. Will go over the list of sites and @kdaily's scripts.
Determine usage of individual synapse visualizations of CSBC sites. 1- talk to @meredithslota to see if it's worthwhile to use google analytics, and how 2- file a jira to get warehouse access 2- Use RMd tool from @kdaily (or improve) to assess views of synapse pages with visualizations
LIst of sites is in here: https://github.com/Sage-Bionetworks/syndccutils/tree/master/R/scripts/csbc_pson
Goal is to finish by August 14th meeting.