emergency-power - #1
Conversation
0a92f1d to
3b9a0cd
Compare
7f5019a to
e41eeee
Compare
stefdegroot
left a comment
There was a problem hiding this comment.
Some small notes, will test further later on, so I might have some more feedback.
| let _ = schedule_tasks(ScheduleGranularity::SECONDS, &[5],app_state.clone(), balance_delta_service, "balance_delta"); | ||
| let _ = schedule_tasks(ScheduleGranularity::MINUTES, &[1, 16, 31, 46],app_state.clone(), merit_order_service, "merit_order"); | ||
| let _ = schedule_tasks(ScheduleGranularity::SECONDS, &[5],app_state.clone(), balance_delta_service, "balance_delta"); //need change to less frequent | ||
| let _ = schedule_tasks(ScheduleGranularity::MINUTES, &[3],app_state.clone(), merit_order_service, "merit_order"); | ||
| let _ = schedule_tasks(ScheduleGranularity::MINUTES, &[5],app_state.clone(), settlement_prices_service, "settlement_prices"); | ||
| let _ = schedule_tasks(ScheduleGranularity::MINUTES, &[1],app_state.clone(), frr_activations_service, "frr_activations"); |
There was a problem hiding this comment.
Merit order updates every 15 minutes, so we fetch it one minute after the source is updated, 4 times an hour.
For the balance delta, it updates every minute, so 5 seconds past the minute we fetch the new value, 60x an hour.
There was a problem hiding this comment.
we had discussed this earlier. I ended up increasing some of the cycle timings based on my specific context. My bad, the idea was tests to overloading and try to make the sync faster.
There was a problem hiding this comment.
I guess you could add 2 api keys to double the sync speed. But not sure if that is the ideal way. I would suggest users download the CSV data set and sync the bulk of the history that way. In the future I want to open up an s3 bucket of end point in our API to expose the full data set.
* basic OpenAPI documentation added * expose openAPI JSON file through the http server * remove unnecessary docker documentation * apply clippy suggestion
b850450 to
73e6273
Compare
Changes: