Take in a JSON log, expand to columns, perform queries.
We can probably pre-define a data set that is useful and then can run queries against it. Or we can download something from S3/github and show against real data. We can probably store this in a folder in this repo as an example dataset too since pip only pulls from what's in the icedb directory in this repo.
Then write a query that shows things like percentile latencies by endpoint, top 1 most hit endpoints, abnormally slow requests.
Take in a JSON log, expand to columns, perform queries.
We can probably pre-define a data set that is useful and then can run queries against it. Or we can download something from S3/github and show against real data. We can probably store this in a folder in this repo as an example dataset too since pip only pulls from what's in the
icedbdirectory in this repo.Then write a query that shows things like percentile latencies by endpoint, top 1 most hit endpoints, abnormally slow requests.