PipelineDB (YC W14) has a lot of similarities to the technology described in this paper, and is open-source: https://github.com/pipelinedb/pipelinedb

Disclaimer: I worked at Truviso; I know the PipelineDB guys but I'm not affiliated with the company.

This is a SIGMOD 2010 paper, the date should appear in the title.

This is based on an older product called Truviso which was acquired by Cisco in 2012.

Shameless plug, but if you want to try a similar product that you can download and play around with today (continuous queries on streams), you can try Striim:

http://www.striim.com

Instead of using Postgres for recovery, the streams are backed by Kafka. You can plug it into your existing streaming infrastructure if you use Kafka today or just deploy Striim standalone.