Henhouse is a fast time series DB that can compute sum, average, and variance between any two time ranges in basically constant time.
DB stores a sums and sum of squares table along with actual values to provide constant time computation between two time ranges.
Henhouse uses the old school embedded signal processing technique of using streaming versions of computing mean and variance.
Read more about the system design: DESIGN.md.
Henhouse is built in C++ and runs on Linux and MacOS. See build instructions for more information.
Henhouse provides both a HTTP query service and a Graphite compatible input service.
You can read about how to use these services here
|docs||Design and build documentation|
|tests||Tests to Hammer Henhouse with Good and Bad Queries|
|tools||Misc Tools to work with Henhouse|
Henhouse was designed and authored by Maxim Khailo (@mempko).
It is currently maintained by Jeff Mataya (@jmataya).
Thanks for considering to help out with our source code! We operate on an open contributor model where anyone across the Internet can help in the form of peer review, testing, and patches.
For more details about how to get involved, see our Contribution Guide.