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
Consider teafiles + lmdb for time series persistence and transactions.
I think a flat file storage system could be a great asset to the project. I have only looked at the teafile spec and have started to use lmdb and Bolt (a golang implementation).
HDF5 seems to be highly regarded in trade apps as well.
I see influxdb in the dev requirements for insight. My problem with influx db for the moment (this could very well change) is that the series have to be sorted after every query it seems.. In a flat file data system we could keep the keys pre sorted by timestamp and write the custom logic to add layers of additional analysis if and when it needs to be persisted with very little overhead to lmdb.
Proof of concept in the works.
The text was updated successfully, but these errors were encountered:
Consider teafiles + lmdb for time series persistence and transactions.
I think a flat file storage system could be a great asset to the project. I have only looked at the teafile spec and have started to use lmdb and Bolt (a golang implementation).
HDF5 seems to be highly regarded in trade apps as well.
I see influxdb in the dev requirements for insight. My problem with influx db for the moment (this could very well change) is that the series have to be sorted after every query it seems.. In a flat file data system we could keep the keys pre sorted by timestamp and write the custom logic to add layers of additional analysis if and when it needs to be persisted with very little overhead to lmdb.
Proof of concept in the works.
The text was updated successfully, but these errors were encountered: