This is good news, unfortunately I wasn't able to attend the UC. The question now is how do I get this to run on premise? My closest Azure data center is in Dublin Ireland which is roughly 9,500km away (as the crow flies). So if I need to replicate the templates across my operations (which I desperately do) then I need to connect all my sites to PICC. The round trip for the data will be somewhere in the region of 20,000km but the distance from the central AF to the operation might be a low as 5km!
It may indeed sound crazy to perform such a long round trip for replicating data between AF servers that are so close. And to be clear, since we are hosting the PI CC infrastructure in Azure, all the data is currently transiting through north American where our solution is deployed. To go back to using as Cloud based solution for local data transfer, I would still strongly encourage you to give it a try. The AF Templates only new publication type is very lightweight as AF Templates information is very small compared to the typical load of a publication containing both AF and real time data. Also, and as we expand the feature set over time, this new publication type will also transfer more AF Objects (Enumeration Sets, AF Analytics and Event Frames templates) that will deliver even more value. We are still looking eventually offering an On-Prem version of PI CC when Microsoft will make the Azure components we aer using available with the future version of Windows Server OS.
As for the UC presentation, you can access the recording at www.osisoft.com/.../item-abstract.aspx .
I definitely will use PI CC for the template replication and more (vendor data). Please don't misinterpret my moan, I think PI CC is a brilliant product with huge potential.
The amount of data for the templates is tiny; probably in the few hundred KB per week. Me trying to develop my own solution would be considerably more expensive and inconvenient compared to sending the data around the world.