A time-series DBMS for R users

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

A time-series DBMS for R users

Leonardo Silvestri

I'd like to announce the alpha release of ztsdb, an open-source
time-series database management system that is very tightly integrated
with R. It should be well-suited for handling financial time-series
data, so potentially of interest to some of you.

The query and manipulation language of ztsdb is a subset of R which
makes it possible to integrate ztsdb seamlessly with R. For example,
from an R session, given a connection 'c1' to a remote ztsdb instance
and a time-series 'z' on that remote instance, one can query the
database from the R command line with unquoted expressions. Here are a
few examples:

## get z's last 6 observations (translates to an 'xts' time series):
c1 ? tail(z)

## get the 2016 subset of 'z':
c1 ? z[as.interval("+2016-01-01 00:00:00 America/New_York -> 2017-01-01
00:00:00 America/New_York-"), ]

## get the 2016 median-minute aggregation of 'z' and assign it to 'mm':
mm <- c1 ? { start <- as.time("2016-01-01 12:00:00 America/New_York")
              end <- `+`(start, as.period("1y"), tz="America/New_York")
              s <- seq(from=start, to=end, by=as.duration("00:01:00"))
              align(z, s, method="median") }

It has a coherent and rich representation of date/time with nanosecond
precision and built-in time-zone awareness.

Additional features include seamless connectivity with other ztsdb
database instances, continuous/streaming updates and C/C++ fast data
append operations.

More information can be found here:
- project website: http://www.ztsdb.org/
- release notes: http://www.ztsdb.org/blog/blog-2016-12-30-first.html
- Gitlab ztsdb project page: https://gitlab.com/lsilvest/ztsdb
- Gitlab R package project page: https://gitlab.com/lsilvest/rztsdb
- Docker repo: https://hub.docker.com/r/lsilvest/ztsdb/


[hidden email] mailing list
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.