Re: Serverless WG and the AppDelivery SIG

Liz Rice

Thanks for the reminder on this Joe.

The Serverless WG has already been performing that role of defining the landscape in their space, identifying gaps, writing white papers... The bit we’re missing is the liaison that we have between TOC and SIGs, which I for one think is valuable. 

After seeing that “Topic 4” diagram I am open to the idea of broadening to SIG App Platform. I foresee some confusion between that and SIG App Delivery with those names, though. Naming is hard. 

I would love to hear thoughts from more folks on the relationship between Serverless and the new Runtime SIG - we could certainly discuss on the next TOC call. Personally I don’t think SIG Runtime is a natural home for CloudEvents, function frameworks and so on, so I think there is a real gap.  

But there is some overlap already: Virtual Kubelet falls into Runtime’s charter, but also appears on the Serverless landscape. Runtime feels like the more natural home; is Serverless really the right place for VK on the landscape?

On 31 Oct 2019, 15:11 +0000, Joe Beda via Lists.Cncf.Io <>, wrote:

Hey fellow TOC members – poke on this?  Thoughts?


A couple of questions in my mind:




From: <cncf-toc@...> on behalf of Doug Davis <dug@...>
Date: Monday, October 21, 2019 at 7:33 PM
To: "cncf-toc@..." <cncf-toc@...>
Subject: [cncf-toc] Serverless WG and the AppDelivery SIG


TOC members,

On a previous TOC call there was a request for feedback from the Serverless WG as to whether or not it would be a good fit for inclusion in the new AppDelivery SIG. During last week's WG call this topic was discussed and the net result is that the WG members felt that while there is some possible overlap, the scope of the WG is different enough that the SIG would not be an appropriate home for the WG. As a concrete example, it seems as though a project like CloudEvents (which came out of the WG's work) would not be in-scope for the AppDelivery SIG. With that, the WG thinks that a separate SIG might be more appropriate.

However, the creation of a Serverless SIG might be too limited in scope. One of the things that we're noticing is that the lines between CaaS, PaaS, FaaS and Serverless are getting blurry. Which means the distinction between the various platforms, or underlying technologies for each *aaS, is becoming less clear. Instead, the differences are becoming more akin to configuration settings rather than entirely distinct platform considerations. As such, the WG felt that an "AppPlatform" SIG might be more appropriate. This SIG could then focus on the underlying hosting technologies leveraged by higher-level SIGs (such as AppDelivery). It would be something along the lines of "Topic 4" in the diagram here:

We can discuss this on a future TOC call, but if the TOC is interested in this we can take the action item to write-up a formal charter to further clarify the proposal that we have in mind.

-Doug Davis/Mark Peek/Ken Owens

Join to automatically receive all group messages.