On 6 Sep 2019, at 23:22, Peter Robinson <pbrobinson@gmail.com> wrote:
As a general rule, in the early stages of an IoT system, when the problem and the solution are not well understood, there is a high rate of deployment of new s/w versions. If these are going all the way to the edge, I’d expect a lot of instability.

I think that I understand the problem better now. Thanks. I’m less sure that I believe the analysis/solution: let’s see what works :-

The other point I forgot to mention (yay jetlag) is that it allows enterprises to deploy App stacks in the same way in the data centre or on the edge, it allows one method of build/test/deploy wherever they run an app. It might start in the DC but they find they need less latency or something so they push it out to the edge networks to get faster responses, I'm seeing a lot of interest in that for ML and data science style stacks.

P

I’ll be tracking that as it’s one of my usecase. However, I don’t, currently, think that it works well as the edge is fundamentally different from the DC, so the testing is different. Unless they’re thinking of something like the AWS Deeplens model.  I can see the value in using containers in a DC (or more partiuclarly a cloud) model. Maybe there are abstractions that make a cloud look like distributed hardware that I’ve not seen yet.

Is there any published work on this approach? (I’d expect that the answer is no)

Tim