Thanks for the question. There are several people from the openwhisk community who are immersed in the knative community, including lots of folks from Red Hat and IBM. I'll trust they will chime in as they see fit.
(Aside: The question has been posed to me often enough that it's worth putting together a blog post on this topic which I will start drafting and circulate to the mailing list for feedback and contributions -- if you're interested in co-authoring this please reach out separately to coordinate.) Back to your question: The short of it is that today they are two separate projects but there are some obvious and some less obvious overlaps. "kwsk" was an proof of concept to show how you can preserve the openwhisk user experience (the CLI programming language) backed entirely by a knative stack. It is - in my belief - mostly just a proof of concept and without some community support to commit to making it real, it is just that: a proof of concept. Knative is worth paying attention to because 1. it originates from Google which has strong OSS marketing muscles, 2. it is riding the kubernetes wave, and 3. does a good job of separating at least 3 dimensions of serverless: building, serving, and eventing. All three of these have direct analogs in openwhisk and it is my impression that a lot of openwhisk concepts exist in knative (whether implicit or explicit). Google doesn't enter markets to make a small ripple. There are also some challenges that knative is addressing and has to address - which others can better speak to because they are immersed addressing these challenges. Some of these are fundamental since if it's tied to kubernetes, then there are fundamental limitations of kubernetes which are not entirely aligned specifically with serverless functions. Other aspects of serverless (just containers) fit more naturally. If knative becomes a defacto standard for serverless on kubernetes, then I see a future where some parts of openwhisk are replaced with those standards. How much I do not know and really depends on how fast knative matures, is adopted, and can address the challenges openwhisk already solves. There is surely more one can say on this topic but I hope this is helpful to your question. I'll add that if one is looking for a production ready, multi-tenant, and fully featured AWS Lambda alternative that's open source, they'd be hard pressed to find a better alternative to OpenWhisk. -r On Sun, Jan 27, 2019 at 11:29 AM Krzysztof Sobkowiak < [email protected]> wrote: > Hi. > > > I've just started my journey with OpenWhisk. I think it's very > interesting. Lately I have heard a > lot about Knative. I thought the technologies are rather competetive. But > I have also seen some > talks where OpenWhisk was placed on top of Knative (on slides). What is > the current relation and > future plans about the both solutions? I have also seen kwsk which looks > like try to integrate both > solutions but nothing happens there since 6 months > > Best regards > Krzysztof > > > > -- > Krzysztof Sobkowiak (@ksobkowiak) > > JEE & OSS Architect, Integration Architect > Apache Software Foundation Member (http://apache.org/) > Apache ServiceMix Committer & PMC Member (http://servicemix.apache.org/) > Senior Delivery Architect @ Capgemini SSC ( > http://www.capgeminisoftware.pl/)
