I'm not sure if anyone here is following along with the elastic 5.x upgrade, but that work is scheduled to happen during the 16.11 cycle which is nearly upon us.
While doing some background checking on where the 5.0 release is (currently in RC status), I stumbled across a new partnership to Elastic which brings with it Machine Learning integration on the data-sets warehoused in Elasticsearch. Now I'm not proposing we immediately hop on board and start charming this up, but it seems like an extension that adds Prometheus style introspection, but with the added benefit of ML on the backend so it learns your model(s) and knows what to expect vs what is an anomaly. http://info.prelert.com/ and http://elastic.co both of information and announcements around the partnership. Background to Prelert: they offer both business security monitoring and operational monitoring for anomaly detection. Is this something we're interested in evaluating as a POC to gather more information? Is someone out there already working on this? It seems to me (assuming this is as awesome as it sounds) that if we were able to ship both long-term metric aggregation and visualization, with an added benefit of anomaly detection, that we'd be offering a powerful 1-2 punch out of the box with our solution. But I don't want to dedicate *any* time to this if we're not even clear if it provides enough value to our potential customers. So I'm seeking commentary of any sort on if this is something we're interested in looking at. Thanks and all the best, Charles Charles Butler <charles.but...@canonical.com> - Juju Charmer Come see the future of modeling your datacenter: http://jujucharms.com
-- Juju mailing list Juju@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/juju
