You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using a longer timeframe for the graph (e.g. 365 days), the request to generate the graph can take a bit longer. This is fine on the initial request, but it would be preferable if consecutive requests are faster.
Requested Feature
Implement caching to ensure consecutive requests are faster. This could clear the oldest entries according to two strategies:
After X amount of time
After N amount of entries are present
You could opt for either, or both of these strategies.
Something to keep in mind is the timeframe parameter. It makes sense to re-use values from e.g. 365 days when requesting just 50 days. The other way around (50 days cached, 365 days requested) would be up for debate. You could use one of the following strategies:
Request data without the past 50 days, so you only have 315 new entries, but would require merging.
Request the full data set and overwrite the existing entry
Personally I'd go for option 2, as it has less overhead, but both have their benefits. Especially if you decide to support larger timeframes (e.g. 5 years) option 1 could be interesting
The text was updated successfully, but these errors were encountered:
Related Problem
When using a longer timeframe for the graph (e.g. 365 days), the request to generate the graph can take a bit longer. This is fine on the initial request, but it would be preferable if consecutive requests are faster.
Requested Feature
Implement caching to ensure consecutive requests are faster. This could clear the oldest entries according to two strategies:
You could opt for either, or both of these strategies.
Something to keep in mind is the timeframe parameter. It makes sense to re-use values from e.g. 365 days when requesting just 50 days. The other way around (50 days cached, 365 days requested) would be up for debate. You could use one of the following strategies:
Personally I'd go for option 2, as it has less overhead, but both have their benefits. Especially if you decide to support larger timeframes (e.g. 5 years) option 1 could be interesting
The text was updated successfully, but these errors were encountered: