Cloud Native Computing Foundation KubeCon + CloudNativeCon North America 2016 (Seattle), 16 Nov 2016

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Keynote: Cloud Native Architectures with an Open Source, Event Driven, Serverless Platform

Description

Keynote: Cloud Native Architectures with an Open Source, Event Driven, Serverless Platform - Daniel Krook, Senior Software Engineer, IBM

New cloud programming models enabled by serverless architectures are emerging, allowing developers to focus more sharply on creating their applications and less on managing their infrastructure. The OpenWhisk project started by IBM provides an open source platform to enable these cloud native, event driven applications.

About Daniel Krook works with customers and the community to create first of a kind cloud solutions based on the OpenStack, Cloud Foundry, Docker, and OpenWhisk open source projects. Most recently he has been building bots and IoT prototypes backed by serverless, event-driven cloud architectures.
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.

A

Hello, my name is Daniel crook I'm, a software engineer at IBM, so today, I'm going to talk to you a little bit about serverless architectures, so serverless architectures have been quite a hot topic in the last year or so because they offer quite a few new technological ways to build cloud cloud native applications and they've introduced a very efficient cost model.

A

So these service architectures get their name because developers, you know with the evolution of cloud, are focusing more and more on the actual applications that they're building rather than the infrastructure. That's underneath them to support them so from the developer point of view their server list, but, of course, there's there's plenty of infrastructure that sits behind them and because the developer is focusing more and more on the application code and rather the infrastructure they provide.

A

The service platforms address many of the operational concerns put forth by the 12 factors: the best practices for developing applications, so developers can get a lot of those benefits and focus on less of the concerns to build applications.

A

Well, another trend: it's making the Popular's there's a bunch of non web use cases that want to take advantage of cloud architectures the scaling models that are in there so such as, if there is a change to an object, storage service or a database change, to provide some logic to handle that event or to handle MQTT messages that come from iot devices to be able to do some edge analytics with that information, conversational BOTS, providing a way to plug into social platforms, a different deployment model and task that may only run for a few seconds or minutes over the course of the day.

A

You don't want to deploy a whole system.

A

You just want to have a time task that runs once a day, for example, and, of course, mobile application backends, where developers focus on really the the user experience of what's in the device, rather than worrying about auto scaling all the backend services and, along with those technical benefits, the developer, the cost model is really providing a new way to map the compute time used by the application more closely with how they're charged so while serverless is not a great fit for every single workload, it's not a silver bullet.

A

It provides a way for these event-driven workloads to have a more efficient cost model, so those technological trends are coming together and making it very compelling and I expect this time next year. There's going to be lots and lots more talk about service architectures, so how did developers actually get to implement service? Serverless architectures? Well, IBM introduced an open source project called open whisk about eight months ago, and this provides a flexible platform for developers to deploy their functions as a service.

A

So it can be run either on a public cloud like bluemix on a local on prem, openstack cloud or the upstream developers who are working on the open source project can deploy open whisk on their laptops. It's also designed to be very flexible, so microservices you want to be able to deploy them in a polyglot way or write them in a polyglot way. So it supports many different languages and a mechanism to extend that with any arbitrary binary through docker.

A

So a developer working with open, whisk they're focused on four key primitives right, they're, defining their event, services, their event, sources, I'm, sorry such as github web hooks IOT data database changes as triggers. They then write applications in Java, Python, JavaScript or docker, which provide basically just implement a main signature which they use to respond to those events, and then they can declare the the mapping many-to-many either way of those event.

A

Sources to the actions that process them and one of the key hallmarks of open, West versus other systems, is that it provides a packaging mechanism. So you can get events not only from the cloud that you're on, but also any arbitrary event source, and these packages can be written by third-party developers. There's a bunch already built in there for integration with with copco msgs, slack IBM watson, services and so on.

A

The easiest way for a developer to kind of kick the tires with it is to go to bluemix dotnet. The value add that we have above the open source project, is a web IDE, some wizards to start with your first actions, as well as the trace, the monitoring of how your actions are doing, how you debug them and what the estimated costs might be.

A

Now there are servers behind the server less architecture, so opens whisk as an open source platform. It's built on a lot of well tested, open source projects such as I'm, sorry, engine X, couch, TV, console kafka and, of course, those actions you write eventually are run on docker containers on a host of nodes behind the scenes you can learn more by following that link.

A

There's a great article written by the core developer on how open west works under the hood, and so we invite you to either try out as a developer on bloom scott net or, if you're interested in joining the open source community we're building a round open whisk, go to github repos at open west org, or talk to developers directly on our public slack channel, and you can also find other IBM open source such as amalgamate a framework for deploying microservices on topic loud, such as Coomer Nettie's act developerworks that open. Thank you very much.