Cloud Foundry Summit Europe 2017

We the Cloud Foundry community have spent a considerable time understanding and unpacking buzzword terms like Agile, Devops, Microservices and Cloud Native. This presentation goes beyond simply unpacking trends and terminology. It will bring you tangible real world examples - from several industries - on how companies have embraced and adopted essential IT trends through leveraging Cloud Foundry to realise substantial increases in profits.

Distributed Service Bundle is a mechanism to define a group of services whose lifecycle and operations are controlled as a single unit. This means parts of the bundle are deployed, un-deployed, updated, operated and monitored together. Cloud Foundry provides well-defined model for application to service communication via the service binding. While this model works well for majority of the use cases, there are use cases where a managed service needs to communicate with other managed services. For example – connecting the Jobs in a managed Spark to Topics in a managed Kafka. Cloud Foundry doesn’t inherently have a notion of service to service binding. To address the above scenarios we need a way to deploy services together with dependencies injected via the environment variables similar to the way CF applications use the VCAP environment variables. In this session, we will explain the need for distributed service bundles and show how such a distributed bundle can be realised via a generic implementation of Cloud Foundry Service Broker.

A superior customer experience is now a competitive advantage, as Amazon so deftly demonstrates through its effective combination of software and data.  We explore how traditional retailers can thrive in this new landscape by delivering a more compelling, personalized experience for their customers.

Starting with personas "modern customer" and "clothing retailer", we demo a microservices-based experimental solution which combines Cloud Foundry, Spring Boot, Spring Cloud Dataflow, cloud-based machine learning, and a mobile app.

The result: better engage with the customer, understand her, and deliver compelling offers that retain her loyalty.

The demo code is available under an Apache 2.0 license: https://github.com/pivotal-cf/oss-pcf-gcp-retail-demo

There are many options when it comes to choosing a cloud platform. How do you know you picked the right one? Each has it's own advantages. What if you want to use more than one? As many of us know, Cloud Foundry provides this multi-cloud portability.

But what about services? There are so many available services and like Pokemon, you want to catch, err use, them all. How do we bring portability to cloud services across industries and across vendors?

In his presentation, Josh will talk about the impact of the Open Service Broker API project. The project provides a simple way to deliver services on cloud native platforms. Cloud Foundry, and Kubernetes are initial adopters. Contributors include individuals from Pivotal, IBM, Google, RedHat, and more.

Many of us have heard the stories of successful Cloud Foundry journeys in enterprises that all begin at the top with a strong sense of shared commitment and vision. These stories can be met with equal measure of appreciation and jealousy by some of us as we feel that it's not exactly a realistic approach in our particular enterprise. This is a different kind of story with different origins, a story of a successful enterprise Cloud Foundry journey with a more complicated path filled with roadblocks, challenge and disbelief. When confronted with shared disbelief like "the world you describe doesn't exist" sometimes the only way to make believers is to build the "fantasy" and let the work speak for itself. Our DevOps transformation and PCF platform leader, Kyle Campos, will share this CSAA Insurance enterprise journey, how we turned skeptics into believers and how we gathered buy in through the doing.

If you've ever deployed your own Cloud Foundry, you may have run the CATs (CF Acceptance Tests) to check that your environment is working correctly. Or, you may have needed to run the CATs in order to submit a PR to CAPI or another core CF team. This talk will present an overview of how these tests are structured, what's needed in order to run them, and how to debug them when they fail. You'll also learn how to add new CATs if you're writing a new component or piece of functionality for Cloud Foundry.

Cloud Foundry is a large and complex ecosystem. Coupled with modern, cloud-native development approaches, the amount of information a developer (or operator) must learn is staggering. Because of this, enterprises face significant challenges when growing teams and capabilities at scale. The old methods simply do not work.This talk will help enterprises understand how to best skill up teams to be effective cloud native developers and/or operators on Cloud Foundry. This talk does not focus on the specific tasks or curriculum, but rather the approach and methodology to building effective organizations. We will discuss the challenge, the shortcomings of traditional approaches, and how to best build lasting and scalable competency in Cloud Foundry and cloud-native.

Gemalto are using Cloud Foundry as the product delivery mechanism for a new marketplace of data protection services. Dan will explain how EngineerBetter have worked with Gemalto to iterate on their platform, build their team and shape their working process. Then Vincenzo will guide the audience through the platform architecture, showing how CF on public cloud has been integrated with physical datacentres to deliver the service. The audience will learn:

• How Gemalto have made unique use of Cloud Foundry’s features to support their go-to-market aims

• Continuous deployment and testing of CF, using Concourse to promote change across multiple environments

• Practical examples of how to build high performance platform teams who have a product mindset

As Cloud Foundry is becoming the Cloud Platform of choice for deploying critical Banking applications, it is essential to understand the architecture and performance of these applications. This session explains the architecture and design choices considered while migrating these applications from on premise to Cloud. This session also covers some of the specific performance challenges encountered and a prescriptive guidance to resolve them which will be applicable for many other Banking applications.

Orange, with more than 263 millions customers worldwide, is a leading company in the telco market.
Within Orange, the “Digital Factory” division particularly focuses on the Orange France marketing needs, with over 40 millions retail customers. It is now offering an on-premise multi data centers OSS Cloudfoundry service. How did we leverage OSS technologies to that end ? How did we first migrate retail applications teams from legacy infrastructure to cloud-ready ? How do we currently operate Cloudfoundry at that scale ? What challenges are we still facing ? Let's join and go together along the path of the business transformation, fuelled up by OSS & community.

In this session Andrea and Fabio will give you a big picture on how a multi-cloud architecture at Swisscom looks like and how to deploy Cloud Foundry on top of it. They'll share their experience and knowledge about multi-cloud BOSH deployments, how to enable continuous deployments within them, where you should pay attention to and - last but not least - how it was engineered. One of the biggest challenges was the connectivity engineering part - at Swisscom we have Cloud Foundry and BOSH deployments stretched over different OpenStack and VMware installations spread across several data centers. This architecture provides a maximum of failure tolerance and high availability as demanded by our customers. We're going to look over the general architecture, operations and continuous deployment of the Swisscom Application Cloud which is based on the open-source Cloud Foundry distribution.You'll also get an overview about how the integration and migration between the different infrastructures can work and take advantage of BOSH's powerful multi-CPI feature and Concourse pipelines to do so. After attending this talk you should know about how to get rid of the most painful challenges in setting up a multi-cloud environment.

Security is a hot topic in Cloud Foundry world, but most CF security conversations revolve around the platform itself – from updating Stemcells to blocking SSH access. What about securing the apps that run on top of CF?

This talk focuses on how you can use Cloud Foundry to build central controls that enforce application security practices on CF apps, resolve issues automatically, and make security controls easy and accessible to apps, allowing them to do the right thing. Investing in such security mechanisms can help prevent problems before they reach production, and continue saving your developers precious time.

Guy will show how the right use of buildpacks, services and external APIs can help handle secret management and rotation, managing vulnerable libraries, encryption handling and more.

In 2015 the Volkswagen Group started with a small team to built an on-premises private Cloud. Today the Group IT Cloud is the base layer of Volkswagens digitalisation strategy. Last year at the european summit Roy and Carsten gave an overview of Volkswagens way to change the companies way to create software. In this talk they will give an overview of the different project phases, what happened meanwhile and how Cloud Foundry helps Volkswagen to become a leading mobility provider.

So you’ve set up Cloud Foundry. Awesome. Your devs have speed, self-service, agility, no downtime. Great! But wait, there's a platform update? How do you know? It's critical? Hurry and download that file, go patch that dev environment! Good? Great! Now patch your other four environments. Nice job! Time for coffee.

Oh wait - there’s another update? Better get back to your workstation…

Patching and upgrading your Cloud Foundry platform, be it the buildpack, a stemcell, a tile, or Ops Manager itself, is still a manual process. An operator must become aware of new updates, download the files from Pivotal Network, and initiate each update through Ops Manager. Rinse and repeat for each of your foundations. Manual processes are inefficient, cumbersome, and often postponed, leading to risk as potentially important updates are delayed or ignored.

Taking a lesson from the app dev community, we can use the principles and practices of Continuous Delivery to alleviate these challenges. Join us to see how you can set up Concourse pipelines to automate the patching and upgrading of Cloud Foundry itself, such that new releases on Pivotal Network trigger the automated application of updates to all of your foundations. Cuz everyone can use an extra coffee break.