What Choices Are Out There for Serverless? (part 5 of 5)


Part 5 of a 5 Part Series. See the other videos here: All About Serverless Computing

So the landscape isn't isn't terribly huge at one level. You've got the major service providers, each with a functions-as-a-service offering that you can put your code into. And you know it's Amazon is Google its Azure and the like. The reason you might go to one of those locations is because of the other services they offer as well. So you can go into Amazon Lambda for example and use DynamoDB and some other offerings they have which are very compatible and growing more so each each week with each release as they enhance their environments. It does create a little bit of a vendor lock-in perspective but that's probably not the worst thing in the world knowing that the world is going that way and that's that's what's going to happen. Your other option is to look at open source versions of container-less. So you know Platform-9 has sponsored a fission project that's F-I-S-S-I-O-N. And there's Apache OpenWhisk which is very popular and actually Apache OpenWhisk is really the thing that IBM has adopted in their cloud and will also bring on premise to you as part of their solution. So that's that's worth taking a look at. Then Iguazio who really focuses on AI and highly scalable data pipelines has now brought forth a project called Nuclio that really pushes the performance edge of serverless and in Jupyter notebooks especially if you're in that AI machine learning world to deploy right to a serverless environment. So you've got a couple of options and I think the biggest thing you have to look at first if you're going this route is open source versus a cloud or in the case of IBM, you can kind of use it both ways and decide from there what to play with.